The dialogue between the Centre and the states has now given a design for the proposed goods and services tax (GST). What has, however, not yet been decided, and seems to be the Gordian knot for the GST is the number of services that should be given to the states.
In principle, the application of GST calls for the same base both under the central goods and services tax (CGST) and state goods and services tax (SGST). It cannot have two different bases, i.e., one for CGST and another for SGST.
This is what has been done in Canada when Quebec wanted to have the right to collect GST, instead of surrendering their power of collection to the federal government through the levy of HST. The federal government of Canada insisted that the base should not only be similar but must be the same. In following this direction, Quebec had to give up some of the exemptions.
A similar principle has been adopted by the central government in the case of CGST. Following this, the proposed CGST would now be levied up to the retail level instead of being confined to the manufacturing level. Although the Centre has applied the principle of keeping the same base for CGST for goods, it has not applied the same principle for services in the case of SGST. As per present indications, all the services would be covered in the base of the CGST but only some services would be included in the SGST.
This can create many problems. First, the services that are with the Centre and the states would have two rates (rate of CGST as also of SGST) to be levied but those services which do not fall in the arena of SGST will have only one rate. The tax rates would, therefore, vary across services in the country.
Second, adjustment for input credit in the SGST for services from the goods would be different for different commodities. Same would be the case with the adjustment for input credit for goods from services. Thirdly, this would create inter-service distortions due to variations in tax rates.
It is, therefore, important that the same base is adopted for services both for the CGST and the SGST. This will not only sort out the issue of whether or not to have the same base but would also give another advantage to the overall system, viz., mobilising larger resources from these taxes. It would allow us to have a lower, revenue-neutral rate for the GST.
Interestingly, calculations from the available data indicate that although 52% of the GDP is contributed by services (excluding public goods and merit goods), 38.5% of GDP could still be included in the taxable base. Similarly, the private final consumption expenditure (PFCE) indicates that 27% of personal expenditure is on services. Excluding expenditure on public goods and merit goods, data indicate that 74% of the expenditure on private services could be the taxable base.
If we add the final consumption expenditure by the government (FCEG) on taxable services to the PFCE, the taxable base would be of the order of Rs 475,000 crore. This could yield India Inc a revenue of Rs 71,000 crore at 15% rate and 95,000 crore at 20% rate of the GST. Looking at the share of this between the CGST and SGST, the former would yield 34,000 crore at 7% and 74,000 crore at 10% rate.
The calculations further indicate that SGST contribution would vary for different states as per capita expenditures differ amongst states.
With taxes at the 10% rate, services would yield revenue above Rs 600 to Rs 856 crore in states of Bihar and Manipur; more than Rs 1,000 to 1,300 crore in Jharkhand, Orissa and Uttar Pradesh; exceeding Rs 1,500 to Rs 1,800 crore in Gujarat, Haryana, Madhya Pradesh, Rajasthan, Tripura and other north-eastern states; more than Rs 2,000 to Rs 2,600 crore in Andhra Pradesh, Assam, Chhattisgarh, Jammu & Kashmir, Karnataka, Mizoram, Tamil Nadu, and West Bengal; more than Rs 2,500 to Rs 2,800 crore in Maharashtra and Punjab; and much more than Rs 3,000 crore in Himachal Pradesh, Kerala and the Union Territories.
The levy of GST on services by the states, however, raises the issue of cross-border transactions. This has been cited as one of the arguments against assigning services to the states. It is said that this, first of all, would involve the issue of who would collect the tax on such services. In this context the report of working group has opined that in a transaction of service where the buyer and seller are located in two different states, the jurisdictional reach of the service tax is to be determined on the basis of place of consumption of service.
This issue, however, is further elaborated in A Model and Road Map for GST in India which proposes that the states may collect tax on intra-state services for CGST and from intra-state transactions for SGST. Similarly, the Centre may collect tax from inter-state services under CGST and from inter-state transactions for SGST. This would sort out the issue of who would collect the tax?
In regard to relocation of funds the empowered committee (EC) is of the opinion that the share of the respective governments would first be decided on the basis of destination principle. It would then be transferred to the respective governments.
Transfer of the central share would not face any problem as the states would directly credit it to the account of the central government. The distribution of the share of the states from the inter-state transactions in services would face some problems. In this context the suggestion of the EC to place this amount in a divisible pool is worth considering. It is proposed that a committee be appointed by the central government to decide on this issue.
It could also be proposed that the share of this pool could be decided by the Finance Commission appointed every fifth year. The Gordian knot of taxation of services would thus be cut by allowing both the tiers of governments (central and state) to tax all the services and not by just transferring 33 services, as was decided earlier. Assignment of a few services to the states does not provide a level playing field.