Next steps for the goods and services tax
Latest News »
The goods and services tax (GST) has been much awaited as a game-changing reform which, if done properly, could add up to 2 percentage points to the growth rate. Unfortunately, the initial enthusiasm has been dampened greatly by the specific proposals put forward by the finance ministry. Unless the GST council, which has to take a final view, modifies these proposals, we will be stuck with what this paper editorially described on Monday as a “flawed GST”.
This article focuses on how the council could retrieve the situation.
Flaws in the present proposal
The proposal before the council departs significantly from what experts regard as an the ideal GST: one with a single rate with very few exemptions. Instead, we have a multiple-rate structure in which about 100 items will be completely exempted (mainly food and other items, broadly those currently exempted from state VAT) while all others will be grouped into four categories attracting GST at 6%, 12%, 18% and 26%, respectively, with a separate cess on the items in the 26% slab to meet the expected cost of compensating states for any revenue loss. Petroleum products, alcohol, electricity and real estate are excluded, though petroleum products are expected to be included within five years.
Multiple rates present problems because a low rate for some goods forces a much higher rate for other goods to ensure overall revenue neutrality. This creates the possibility of misdeclaration of goods to benefit from the lower rate. This problem can be minimized if all items that are similar are taxed at the same rate. However, a second problem remains: High rates of duty encourage evasion. Experts favour a moderate common rate applied on all goods precisely because it is expected to encourage compliance, which is a major reason for moving to a GST.
International experience is relevant in this context. About 50 countries, including many developing countries, have switched to some form of value-added taxation in the past two decades and 80% of them have introduced only one rate, with some having two or more. The European Union has multiple rates, but they have recently released a Green Paper for discussion on how to converge to a single rate. Should we not leapfrog to best practice instead of trudging laboriously through paths others have trudged and then abandoned?
Policymakers like to have low rates on some commodities because they want the tax system to be progressive. But progressivity does not require different tax rates for too many commodities. If the exempted category forms a large proportion of the consumption of lower income groups, it will ensure that the total tax paid as a percentage of total consumption will be much lower for lower income groups. Besides, this approach to evaluating the impact of the GST on the poor is incomplete because it gives no weight to the efficiency gains which are expected to produce higher growth, higher employment and higher wages.
This is not to suggest that there will be no gains at all from what is being proposed. The fact that the Centre and the states will both be levying taxes on the same base across the whole country, with both going up to the retail level, is clearly a gain in terms of broadening the base and making tax administration simpler. So is the abolition of the central sales tax (CST) in terms of unifying the market. The levy of a countervailing duty (CVD) on the import of a commodity at the same rate as the GST is also an important gain that will give domestic producers a level-playing field versus imported goods, something that has been missing for many years. The real problem is that the departure from an ideal GST will mean that productivity gains for the economy will be much less.
Vijay Kelkar, former chairman of the Thirteenth Finance Commission, which recommended how we might move to a GST, has recently stated (see his V. Sankar Aiyar lecture in Chennai) that the multiple-rate structure proposed, with numerous exemptions, will reduce the growth gains to one-fourth of what they might otherwise have been! In other words, instead of the extra 2 percentage points in GDP growth that has been talked about, we might gain only half a percentage point. The GST council must ponder whether we can throw away 1.5 percentage points of growth.
What can we do?
The first best solution is to follow the advice of Kelkar, Satya Poddar and V. Bhaskar, in their article published in this newspaper on 19 October, in which they recommended a common rate with an effort to offset the adverse impact on the poor through a direct benefit transfer (DBT). They estimate that a transfer of Rs2,000 per person per year would more than suffice to offset any increase in post-tax prices and would cost only Rs50,000 crore per year, which is a small price for a perfect tax. This is a neat solution which will have the added advantage that it will test out the effectiveness of DBTs, which could then be used more broadly to phase out other subsidies.
However, the best must not become the enemy of the good and the GST council may not be able to deliver the best. It may have to settle for the good, but in that case it must ensure that the good is good enough. If a single rate is not possible, the GST council could at least improve the proposal by reducing the number of rates immediately and announcing the intention to converge to a two-rate structure in the near future. This can be done as follows.
We should do away with the proposed 6% rate and move all these items into the 12% rate slab. The present list of items could bear a higher rate. If it is felt some of the items in the list are too sensitive, they could even be exempted in exchange for all the others going into the 12% group.
At the upper end we should do away with the proposed cess on items attracting 26%, which in any case violates the basic principle that all cesses would be subsumed in the GST. Instead, we should add an element of “sin tax” to the 18% rate to reflect a higher rate on “luxury consumption” and undesirable consumption (e.g. tobacco and also fossil fuels) and provide that this additional tax should not get credit in the GST chain.
This would give a three-rate structure: 12%, 18% and a third rate which is 18% plus sin tax. Many of the items in the present 26% slab should be included in the 18% rate slab while others that are in the nature of luxury items could be in the 18% plus sin tax slab.
The GST council could further announce that it envisages a convergence to two rates in future, with the manner of convergence to be decided later on the basis of experience in the first three years. The process of convergence could begin on 1 April 2020, after the general election.
Experience shows that small adjustments in prices are more easily accepted than large increases and we should capitalize on this in designing the convergence formula. For example, if experience in the next three years is positive, and states realize more revenue than they currently expect, one can imagine the GST council endorsing a proposal for convergence over three years by raising the 12% rate by 1 percentage point a year and similarly lowering the 18% rate by 1 percentage point. If this process begins in 2020, we would end up with a single standard rate around 15% by 2023, with a second higher rate including sin tax. Hopefully by then petroleum products will also be included in the GST, and real estate too.
The GST council is an important new institution of governance in our federal system. A recommendation from this body is perhaps the only way of depoliticizing the process of tax reforms, which is essential if the Centre and the states are to take bold decisions in this area.
An independent secretariat for the GST council
To accomplish this ambitious task, the GST council needs to be supported by an independent secretariat with technical expertise. State governments have a two-thirds vote share in the council and are therefore critical partners in decision making. However, most state governments lack the technical capacity to evaluate complex issues related to the design of indirect tax structures, especially since we have to consider the impact on the country as a whole and not just on individual states. The finance ministry is much better placed in this respect, and it can also draw on the expertise of the National Institute of Public Finance and Policy. However, the states are unlikely to accept finance ministry inputs as being sufficiently independent.
It is, therefore, necessary to set up a high-quality independent technical secretariat to service the council. The secretariat should be separate from the finance ministry and should report to the chairman of the council (i.e., the finance minister, but in his capacity as chairman and not as the finance minister). The unit should respond to concerns of state governments on the impact on the states of specific proposals. If the finance ministry wishes to make submissions to the council, these should be evaluated by the independent secretariat and the evaluation circulated to all members. The secretariat should also produce or commission technical papers that can guide future evolution of the rate structure and these should be put in the public domain. This would also enable business and industry groups to lobby their point of view with both the Central government and state governments.
The GST council has very limited time given the 1 April deadline. It cannot sort out all the problems in the short time available, which means that whatever is approved now will have some flaws. However, these flaws will attract less criticism if the council clearly states a longer-term objective of converging to two rates and establishes a mechanism for considering ways of getting there. This will certainly restore confidence in the seriousness of our commitment to reforms.
Montek S. Ahluwalia was the deputy chairman of the erstwhile Planning Commission.
Comments are welcome at email@example.com