1 Introduction

One of the most contentious issues in the discussions surrounding GST is the likely and feasible rates at which the new regime can be implemented. Under this broad rubric, there are two broad issues that emerge. One, “very high rates” of tax, it is argued, will not encourage compliance. There are various notions of what constitutes “very high”. The Kelkar Committee Report had proposed a figure of 20%, and this has become one benchmark for what constitutes limits of an acceptable rate. Second, at whatever be the chosen rates, would all the states and the central government get “adequate” revenue? Here adequate has meant equivalent to that from replacing the existing taxes levied by the states. However, there are no such clear definitions of “adequate” espoused by the central government. Since the central government, in a sense, is the champion of this reform process, the final cost of adjustment, it is assumed, would lie with it. In other words, there are two choices with the central government – one, let the states choose rates that ensure revenue neutrality for each state individually, or negotiate a common rate, with an assurance of a compensation in the event of a shortfall in collection.

In the process of transition to state value added tax (VAT), the central government had exercised the second option and provided a cushion in terms of compensation for revenue shortfall.

Economic & Political Weekly

EPW

SPECIAL ARTICLE

In the present scenario too, there is a discussion of compensation from central resources – the Thirteenth Finance Commission has been assigned the task of factoring in the impact of the introduction of GST while firming up its recommendations. In this context, it is important to understand the size of the GST base in India, within the limits of the design as it emerges from the discussion paper and the likely differentials in the rate of tax that would prove revenue neutral for different states.

There have been a number of attempts at estimating the size of the tax base and the corresponding revenue neutral rate. The latest in the series is the report of the Task Force on GST of the Thirteenth Finance Commission. Most of these exercises throw up incredibly low revenue neutral rates causing apprehension about the validity of these estimates and consequent revenue risk.

This paper seeks to estimate the base for GST in India on conservative assumptions to arrive at more realistic estimate of the revenue neutral rates across states. This is preceded by a discussion of the proposed design of GST in India as laid out in the discussion paper. Some concluding remarks are presented in the last section.

2 Overview of the Discussion Paper on GST

The key features of the proposed GST as laid down in the discussion paper provide a very interesting and useful first reference for any discussion of GST in India. As a beginning, the design proposed in the paper provides considerable detail and structure to the new regime. The key features can be summarised as follows:

• Dual GST: A central and state GST to apply on the same base. Each tax would operate on the principles of input tax credit, where tax credit for each tax would be self-contained. No cross-credit would be allowed except in the case of interstate transactions.

– For interstate transactions, the discussion paper proposes a regime called IGST, wherein the exporter charges IGST in place of the GST in such transactions, claims input tax credit for any taxes paid, and remits the balance to the central government. The exporting state remits any local taxes that are claimed as credit by the exporter to the central government/central administrator. In the importing state, the importing dealer is allowed to claim tax credit for the IGST paid from the central government. The claims across states will be cleared by a central clearing house kind of mechanism. It is proposed to cover both B-B and B-C transactions.

• Uniform state GST threshold is desirable – proposed at Rs 10 lakh.

• Rate Structure: Apart from exemptions for all commodities presently exempt in the state VAT system, it is proposed to have two rates – a lower rate for “necessary items and goods of basic importance” and a standard rate. It is suggested that a similar approach be adopted by the centre for taxation of goods. For services, however, a single rate is proposed. It is also proposed that all exports out of the country will be zero-rated and all

50 imports into the country would be subject to GST, with credit being made available in subsequent transactions.

• Central and state taxes to be subsumed in GST include central excise, additional excise duties, service tax, additional customs duty, special additional duty of customs, surcharges and cesses, for the centre and state VAT, entertainment tax, luxury tax, betting taxes, state cesses and surcharges on supply of goods and entry tax not in lieu of octroi.

• Multiple Statutes: One for the centre and one each for the states. The discussion paper proposes that the “basic features of law such as chargeability, definition of taxable event and taxable person, measure of levy including valuation provisions, basis of classification, etc, would be uniform across these statutes as far as practicable”.

• Tax administration by respective tax departments

Table 1: What Is Service Sector as Per NIC Classification?

EPW

SPECIAL ARTICLE

– “Keeping in mind the need of taxpayer’s convenience, functions such as assessment, enforcement, scrutiny and audit would be undertaken by the authority which is collecting the tax, with information sharing between the centre and the states.”

Some Observations on the Proposed Design

Since this is a discussion paper, with the decisions not completely frozen yet, it may be useful to highlight some of the limitations of the proposed design. In the following discussion, an attempt is made to identify some of the key features which might benefit from a re-examination.

(1) Rate Structure: The rationale for the proposed rate structure is reasonably clear. However, it is important to define what constitutes a necessity or a good of basic importance. In the present VAT regime, a number of inputs were classified in the lower rate, which, as have been often argued does not deliver any benefits in a VAT regime. In the classification of commodities, therefore, it would be useful to limit the coverage of goods in the lower category to a bare minimum. In other words, adopt a conservative definition of “necessity”.

• Apart from this dimension, it is not exceptions are electricity duty and passenger and goods tax at the state level. As stated in the discussion paper, “In the GST, both the cascading effects of CENVAT and service tax are removed with set-off and a continuous chain of set-off…is established which reduces the burden of all cascading effects. This is the essence of GST and this is why GST is not simply VAT plus service tax.” However, retaining these two levies outside GST and keeping crude, and other petroleum taxes outside the base for GST contributes to cascading in the system. It is useful to ask whether, the inclusion of any of these levies into the base would alter the “revenue neutral rates” substantially. This could be one reason for choosing this approach. The following section attempts to answer this question, at least so far as electricity duty and passenger and goods tax are concerned.

(4) Proposed Thresholds: While it is not clear whether the discussion paper is proposing a design for both central and state GSTs or not, there is no spelt out rationale for having separate exemption thresholds for these taxes. A higher exemption threshold for the central GST would imply a significant cost of transition from one rate category to another for any taxpayer.

For the thresholds discussed for instance,

Table 2: Service Sector Sales as Per NIC Classification

clear, why the discussion paper espouses (in Rs crore) at Rs 1.5 crore turnover, there would be a

Sales Data: As Per NIC Code 2007-08

the cause of two rates for the central GST sharp increase in the rate from only SGST

Service sector sales (all-India) 3,35,189

as well. The need for special treatment to SGST+CGST. This could encourage

Revised sales figure 8,44,986

of “necessities” is ensured by one of the units to under-report turnover at this

GDP at factor cost 43,20,892

taxes adopting a preferential treatment threshold or to split a firm into two parts

Service sector GDP 23,39,468

approach. By allowing the other tax to so as to avoid the central tax. The sheer

Service sector sale as % of GDP 19.6

be neutral among these commodities, the fact that these taxpayers would be part of

Service sector sale as % of service sector GDP 36.1 incentive of the taxpayer to comply with Service sector excluding exempt sectors 12,27,470 the SGST system implies that they would the regime can be augmented. Sharp dif-SS sales as % of services excl exempt services 68.8 have the accounting keeping and other

ferences between the rate of tax applicable to these two categories of commodities could discourage compliance at the margin.

• In the treatment of services, while a single rate is proposed, it is not clear whether this rate would be aligned to the standard rate for goods or not. Choosing a different rate for taxation of services would encourage misclassification at the margin without delivering additional benefits. It would be useful and simple for an alignment of the standard rate for goods with the rate applicable to services.

(2) Treatment of Interstate Services: While it is recognised in most of the discussions on GST that a clear definition of treatment of services spanning more than one state is essential before the states can be assigned a comprehensive power to tax services, the discussion paper does not provide enough inputs on this subject. Since most of these services contribute substantial amount of revenue in the present service tax, it would be imperative in the interest of clarity to spell out the proposed treatment of such services, especially for the service providers since they would need to realign their administrative systems to account for a substantially altered tax and accounting regime.

(3) Taxes to be Subsumed: While a number of the important taxes are proposed to be subsumed into GST, two important

Economic & Political Weekly

EPW

january 2, 2010 vol xlv no 1

such compliance features in place. The higher threshold for the central tax would therefore be tax incentive for small businesses. Given the overall decision to keep the tax regime free of tax incentives, it is useful to evaluate whether the central tax should have a higher exemption threshold of Rs 1.5 crore for goods and a completely different threshold for the taxation of services.

(5) Administration of the Tax: The discussion paper emphasises at a number of places that there is need for harmonisation of the forms and procedures, and the taxing provisions across the different tax administrations. This view is expressed in the context of the formulation of the tax statutes as well as in the context of forms and procedures. While this a commendable position to take, and would allow for considerable ease in compliance, once an agreement of this kind has been possible, it is tempting to ask whether one cannot go a step further and explore the possibility of a single registration and a single return. This process need not undermine the administrative independence of the tax departments, and yet could provide a significant simplification for the taxpayer. Given that the tax departments are proposing a common PAN-linked taxpayer identification number, such a process along with a common information system would remove “sharing of information” from a task to a fact and thereby improving tax administration significantly.

SPECIAL ARTICLE

Table 3: Input Coefficients and the Adjusted Base
	Taxable	Share of Sales	Net Ad-
	Inputs/Output	Used as	ditional Base
	Ratio	Inputs	(Rs Crore)
Hotel and restaurant services	0.310535	0.137036	5,906
Transport services	0.407863	0.43389334	41,279
Post and telecommunication services	0.124205	0.70641745	33,211
Financial services	0.07555	0.72606917	39,070
Real estate activities	0.046972	0.06472144	23,187
Renting of machinery and equipment without
operator and of personal and household goods	0.015183	0.64301983	4,679
Computer and related activities	0.103611	0.07353309	86,181
Research and development	0.045501	0.00088177	54
Other business activities	0.127913	0.84847271	21,475
Education	0.127913	0.00088177	15,712
Health and social work	0.299092	0.00893679	474
Sewage and refuse disposal, sanitation
and similar activities
Activities of religious organisations and other
membership organisations	0.13176031	0.41057617	868
Recreational, cultural and sporting activities	0.13176031	0.41057617	6,282
Total			2,78,384
Additional services coming under tax net
Railways: Passenger earnings	11,750
Railways: Freight on exempt commodities	2,471
Air fare	20,606
Electricity value added – domestic consumption	23,540
Electricity value added – agriculture consumption	31,735
Services excluded in the estimation of base
Real estate activities	23,187
Computer and related activities	86,181
Net additional base for GST	2,59,119

Source: Derived by authors using Input Output Tables, Prowess database and service tax collections.

Apart from the pure design of the tax, there are two important issues raised by the discussion paper – one, relating the mechanism for upholding the harmonious structure being proposed and, second, for compensation of current revenue from central sales tax (CST) on a permanent basis. It would be useful to explore what each of these issues suggests.

In the context of the first issue, the discussion paper suggests that

“an appropriate mechanism that will be binding on both the centre and the states would be worked out whereby the harmonious rate structure could be upheld, if necessary with a collectively agreed upon constitutional amendment”.

It is far from clear from this statement what form the mechanism is expected to take. States in India have been strongly vociferous on issues of fiscal autonomy. At one extreme, the above proposal could mean an agreement on uniformity which gets enshrined in the Constitution. At the other extreme, it could require the setting up of an institution which can monitor and limit the scope for variation across states. In the regime of state VAT, the central compensation package provided the tool for ensuring a degree of compliance with the agreed structure. However, since compensation packages are likely to be limited to a few years, the discussion paper proposes to explore some alternative means of ensuring compliance. It is important to recognise that any measure proposed would limit the autonomy of the states and possibly that of the centre as well. A conscious decision can, however, still be encouraged in the interest of simplicity and integration of the country into a single common market.

The second contentious issue in the discussion paper relates to the compensation for loss of CST revenue. The discussion paper states that “the rate of CST has already been reduced to 2% and will be phased out with effect from the date of introduction of GST on the basis of such GST structure which, with necessary financial support to the states, should adequately compensate for the loss of the states on a permanent basis.”

There are potentially two views on this topic: one, since CST is an origin-based tax, it resulted in a flow of revenue from consumers in one state to the exchequer of another state. Since the former do not derive any benefits from the services provided by the latter, such a tax is unfair. Compensation for such a tax, that too on a permanent basis, therefore would be unfair. The second view would be that since some states have been extensively dependent on this source of revenue, it would be unfair to propose a drastic reorganisation of the tax powers, thereby limiting the revenues from such a source.

It should be mentioned here that the expanded base in GST potentially does provide a cover for the erosion in base as a result of elimination of CST. This, however, does not apply symmetrically across all states. Mineral-rich states, for instance, could lose revenue from CST without adequate gains from expansion in base from services. Whether this situation calls for a sustained compensation package or a change in some other parameter, such as royalty on minerals, needs to be explored in some depth in attempting to find a solution to this problem.

3 Assessment of the Tax Base and Revenue Neutral Rates

We have taken data for the fiscal year 2007-08 for the purpose of RNR calculations, since that is the latest year for which comprehensive data is available. For the purposes of this exercise, we have defined the revenue target as equiva-

lent to that raised through VAT, CST and electric-

Due to GST: Major States (Rs crore)

Andhra Pradesh 19,729

Bihar 8,824 Chhattisgarh 3,206

information on the bases associated with these Goa 917 taxes is not accessible, they are kept outside Gujarat 14,877 the exercise. Haryana 7,414 It may be mentioned that since these taxes Kerala 13,849 come with some associated bases, it would not

Karnataka 15,667

Madhya Pradesh 9,570

Maharashtra 43,923

Orissa 5,397

Punjab 8,110

Rajasthan 9,308

Tamil Nadu 21,790

Uttar Pradesh 21,685

West Bengal 21,934

Total 2,26,200 Source: Computed. associated with state VAT is not readily available,

the information on revenues was used to derive an estimate for For all the services, two kinds of adjustments have been made, the taxable base, based on the following assumptions. Thirty per viz, deduction for taxable inputs used for service provision and cent of the trade tax revenue has been deducted on account of deduction of services provided when used as inputs into taxable revenue from petrol, diesel and alcohol. For the rest of the reve-activities. For these corrections, the input-output table for nue, 2% of total taxable turnover is assumed to be taxes at 1%, 2006-07 has been used to derive service specific input-output 33% of the taxable turnover at 4% tax rate and the rest at the ratios (Table 3, p 52). As evident from Table 2, service sector sale standard rate of 12.5%. as a percentage of service sector GDP excluding

Table 5: The Revenue Neutral Rates

the exempted sector. are no or fewer commodities in the 4% category, the RNR as

Sales data obtained from the Prowess data set is a very con-shown in Table 3 can be considerably lower. If, for instance, servative estimate of service sector activity. We have tried to have the tax base subject to 4% rate is limited to 20% of the taxable a more realistic estimate of the service sector base by mapping turnover of goods instead of 33%, the revenue neutral rate service-wise tax collection data against the data on sales obtained can be brought down by over a percentage point for each of the as per the NIC code and adjusting the turnover/sales based on states considered. the tax collected from each category of services. Having done As mentioned earlier, having GST without some of the importhis adjustment, we have obtained a revised sales figure for the tant services being subsumed may result in cascading and service sector at Rs 7,54,883 crore. This is further adjusted for also the real benefit of GST as a non-distortionary tax without cassome of the important currently exempt services which would cading would be substantially compromised. Thus, it is important be part of the GST base, viz, railway passenger fare and railway to find out if inclusion of both goods and passenger tax and the freight on exempt commodities, electricity revenues from sup-electricity duty would alter the revenue neutral rates to a signifiplies to domestic consumers and exempt sectors and air fares. cant extent. We have reworked the revenue neutral rates when