Some Concerns Regarding the Goods and Services Tax
The proposed Goods and Services Tax would aggravate the already high degree of inequality in the country, and would inevitably curb the fiscal autonomy of state governments significantly. Contrary to the government’s claims, the implementation of the GST would not automatically enhance growth rate, reduce inflation or improve tax compliance. The calculations of the revenue neutral rates depend on various assumptions. Different states should be allowed to set their own state–GST rates with provisions for an entry tax. More emphasis should be put on direct taxes rather than the GST for improving India’s abysmally low tax to gross domestic product ratio.
The goods and services tax (GST) is nothing but an ad valorem sales tax on all goods and services with input tax credit. According to the current proposal, we understand that while there would be a variation in the GST rates for different commodities or service groups, the tax rate for any particular commodity or service would be uniform across all the states. Like the present value added tax (VAT) system, the input tax credit ensures that the producer does not pay tax on the material input cost of production and only on the value addition. Therefore, there is no question of payment of tax on the already paid tax amount on inputs, what is called the “cascading effect”.
Almost all other indirect taxes—except custom duty collected by the central government, and duty on petroleum, electricity, and alcohol, levied by the state governments—are supposed to be subsumed under the GST to simplify the tax system and to relieve producers and suppliers from paying multiple indirect taxes. Both the central and the state governments would levy GST on the same tax-base on all the transactions of commodities and services and there would be two rates, the Central Goods and Service Tax (CGST) and the State Goods and Service Tax (SGST) respectively, for an understanding of revenue sharing.
The final consumers would have to pay the entire tax amount at the combined rate, the Integrated Goods and Service Tax (IGST). The rates have to be decided by a centrally empowered authority—the GST council—constituted by representatives from the states, and headed by the union finance minister. The council will ensure that the revenue receipts as proportion to the respective tax-bases (sales) do not decline for the central and the state governments, that is, the GST rates should be at least made revenue neutral. This tax reform has rightly been described by the government as “the single biggest tax reform measure undertaken since independence to usher in a new era”.
The following main claims have been made by the present and past central governments of National Democratic Alliance (NDA) and United Progressive Alliance (UPA) regarding the GST. It is to:
(i) enhance growth and reduce inflation,
(ii) shift the tax burden from producers and suppliers to the final consumers,
(iii) integrate the national market,
(iv) improve tax compliance,
(v) mitigate the cascading effect of taxation,
(vi) be a win-win situation in terms of revenue for both centre and the states,
(vii) be good for both producers and consumers,
(viii) be beneficial for both tax payers and the tax administration,
(ix) make for a simpler and more transparent tax system and so on.
Is everything necessarily rosy about the GST? Are there trade-offs involved in the process? Are questions about the GST irrelevant, now that the constitution amendment bill has already been passed almost unanimously in both the houses of the Parliament, and now that many state governments are supposedly “on board”? Concerns are not necessarily expressed to oppose reform but to engage in serious deliberations. The trade-offs, if any, must be made explicit to enable the designing of a better tax policy for the nation in the future.
Indirect Taxation and Degree of Inequality
Let us take a look at the broad picture of taxation in India. The data on the combined tax revenue of the central and the state governments during the last ten years[i] shows that the combined tax revenue of the central and the state governments in India is around 16.5% of the gross domestic product (GDP), which is one of the lowest in the world (Das 2014a). At present, one-third of our total tax revenue is collected from direct taxes (5.5% of GDP) and the two-thirds are collected from various indirect taxes, which makes for 11% of GDP). It is worth mentioning in this context that, 60 years ago, when the proportion of indirect tax was around 70% of the total tax revenue, I S Gulati expressed deep concern about the extremely high dependence on indirect taxes in India. Nicholas Kaldor (1959) also emphasised the vitality of having an effective system of progressive direct taxation for the survival of democratic institutions in India.
Yet, our dependence on indirect tax is still as high as 67% today. According to the present understanding of fiscal federalism in India, the central government collects more than 97% of direct tax revenue, while the state governments collect around 55% of the total indirect tax revenue. Again, around 97% of the states’ own tax revenue comes from various indirect taxes. Following the recommendations made by the successive finance commissions, after transferring the states’ share, out of every single rupee of tax collection, the centre gets 46 paise and the states get 54 paise. It is important to note here that the division of expenditure responsibilities between the centre and the states is not going to change with the GST. The design of direct tax in the country would also remain unaltered. The introduction of GST will only lead to changes in the structure of indirect tax.
One obvious trade-off involved with GST is that any attempt to increase the tax–GDP ratio through GST would lead to higher inequality in a country which is already witness to extremely high and increasing inequality. GST is an indirect tax and is levied at the same rate on the poor and the rich. Following Kalecki (1943), if the poor consume proportionately more out of their income, the incidence of an indirect tax would be higher on the poor than the rich, which makes it regressive taxation. Given, any degree of progressiveness in direct taxation, if the proportion of indirect tax collection increases relative to direct tax revenue in the economy, clearly, the degree of inequality would rise. Therefore, given any aggregate direct tax to GDP ratio and for any given composition of combined government spending, if the indirect tax collection as proportion of GDP somehow increases under the GST regime— either by better compliance or higher GST rate, or because of larger tax base—the degree of inequality would rise in the economy. If not, then the total tax revenue would not increase as a proportion of aggregate national income. If this abysmally low tax–GDP ratio does not increase, then, given the rule-based budgeting act, or the Fiscal Responsibility and Budget Management (FRBM) Act, the government’s expenditure as proportion of GDP would also remain low.
On the Question of States’ Autonomy
Another obvious trade-off is related to the question of making tax rates uniform across the states and the fiscal autonomy of the states. Under the present scheme of GST, even if the states do not make any revenue loss, individual state governments would not be able to decide the tax rate on any particular commodity or service autonomously, which would be decided centrally by the proposed GST council. Even if the states jointly have some say in the GST council, no single state would have the freedom to decide a different tax rate on a particular good or service.
If the entry tax is abolished, effectively no state would be able to maintain differential tax rates, anyway, for all practical purposes. If the states do not have the power to decide the tax rates, they would become mere tax collecting agents of their own taxes who get a commission depending on the size of their respective tax bases, and receive rewards for “efficiency” in tax collection. Moreover, since a significantly large proportion of the state’s own tax revenue collection comes from indirect taxes, given any vertical devolution by the central finance commission, the total expenditure of the state government as a proportion of respective gross state domestic product (GSDP) would also largely be decided centrally for any cap on the fiscal deficit to GSDP ratio of the states. This again would be decided by the central government or by the Reserve Bank of India (RBI).
This trade-off has been recognised by economists across the board, from the late Amaresh Bagchi (2006, 2007) to Prabhat Patnaik (2016a), Govinda Rao (2009, 2011) and to Kavita Rao (2010). The rationale for having uniform tax rates in order to harmonise the tax base or for a harmonised tax system is not very clear. Apparently, it is desired by the businesses (Govinda Rao 2009). Even if the tax rates are made uniform across states, the price of the same commodity would vary from state to state and place to place, and from rural to urban areas. The price would depend on the cost of production, transportation, marketing, and on the demand, mark–up, etc. The variation in tax rates only explains a small fraction of the total price variation. Hence, the slogan of “one country, one market, one tax rate” may not necessarily hold even as it sounds catchy. The question is if the need for uniform tax rates is pressing enough to prioritise it over the autonomy of states.
Patnaik (2016a) has pointed out that the United States (US) does not have a uniform VAT like the GST. Different states in the US tax commodities by value, not value added, at different rates which, moreover, vary across commodities within each state. In terms of population, India is larger than the entire of Europe and the population of any of its bigger states is higher than that of the United Kingdom, Germany, France, Italy, Spain and so on. The European Union (EU) region has differential rates of GST and it varies from country to country above a certain minimum floor rate set for each commodity/service (Bagchi 2006, Kavita Rao 2010).
We must remember that the division of states in independent India initially took place on the basis of linguistic diversity and that our sub-national identities are extremely strong. Given socio-economic, political, historical, cultural and linguistic diversity, India can remain united as a federal democracy only if the autonomy of states is valued. Our constitution recognised this importance and accordingly made provisions for the autonomy of states in revenue mobilisation. Amaresh Bagchi (2006) has argued that the essence of taxing power lies in the power to fix rates, and thus subnational governments must have the autonomy in matters of fixing the rates, subject only to a floor, as in the EU. Bagchi (2007) mentions that in New Zealand, where the GST was introduced in 1986, the trade-off involved in choosing between the imperatives of fostering a common market and the need to preserve the fiscal autonomy of the subnational governments was historically evident. Once, GST is implemented following a particular design, it will be very difficult to alter it. If the SGST rates are not allowed to vary from state to state with some provision for entry tax, the fiscal autonomy of the states might be compromised forever.
Shifting Tax burden from Producers to Consumers
It is being claimed that under the GST, the tax burden would shift from producers to the final consumers since it marks a shift from origin-based taxation to destination-based taxation. Would this shift from excise duty to ad valorem tax for the “ease of doing business”, necessarily ensure that the economic incidence of taxation only falls upon the final consumers and not on the producers and suppliers? Undergraduate text books of public finance (Musgrave 1959) tell us that irrespective of the degree of monopoly or competition, under a per unit excise duty or an ad valorem tax like the GST, both producers and consumers would have to necessarily share the tax burden given any negatively sloped demand curve (in price-quantity plane), and given the non-decreasing (increasing or constant) cost schedule. In both the cases, the price would increase and the quantity produced would come down for profit-maximising firms, and the shares of tax burden would depend on the slopes of the respective demand and supply schedules for any particular industry. It is not necessary that the tax incidence on producers will be relatively lesser with the ad valorem tax under the GST.
Growth and Inflation under the GST Regime
Insights from microeconomics also tell us that an ad valorem tax is relatively more efficient than the excise duty in the sense that for raising the same amount of tax revenue, the required price rise and reduction in production is relatively less for the former (Musgrave 1959). However, if a higher level of tax revenue has to be raised from an ad valorem tax, price reduction and quantity enhancement are not inevitable. Moreover, if the effective IGST rate is higher than the revenue neutral rate (RNR), the real disposable income of people would come down, which in turn, may shift the demand curve further down towards the left. This means that, given any (positively sloped or horizontal) supply curve, the level of equilibrium output would be even lower, given the profit maximising behaviour of the firms.
Given our tax information network and efficiency of the tax administration, the combined indirect tax revenue for the states and the centre would go up as proportion of GDP only if the weighted average IGST rate is higher than the RNR. The IGST rate has to be higher than the RNR also to ensure net revenue gains for the centre and at least for some states, and no losses (after compensation, if required) for the others. At least for the next five years, if the centre’s net revenue gain after compensating for the losses of some states becomes negative, it would have serious implications for tax devolution and for central government expenditures, given the FRBM Act. Therefore, the combined indirect tax revenue (as percentage of GDP) has to go up under the GST regime, which does not ensure higher growth or lower inflation.
From the point of view of macroeconomics, any effective GST rate above the RNR would cause a reduction in the propensity to consume at an aggregate level, followed by the redistribution of real income (Kalecki 1943) against the relatively poorer section of masses (increased inequality). This would happen because the propensity to consume in general is higher for the poorer people than for the rich. The multiplier (Keynes–Kahn) would be weakened and would result in a relatively less increase in aggregate output and employment, (at a given level of technology), due to the same increase in autonomous investment demand or export demand or government expenditure. The aggregate saving may go up through the generation of the “forced saving”. However, in absence of sufficient aggregate demand in the economy, there is absolutely no reason to believe that the extra saving would necessarily be invested to enhance the aggregate level of output and employment.
Since, at the aggregate level, the RNR of GST cannot be lower than the sum of the rates of taxes that are being subsumed, there is no question of reduction in the prices. What is more, due to the absence of “cascading effect”, the RNR has to be even higher for compensating for the revenue loss. In fact, as argued earlier, the weighted IGST rate has to be higher than the RNR. Therefore, the average price level is likely to increase and not reduce under the GST regime, which in turn, would aggravate inequality. As a consequence, given negatively sloped aggregate demand curve, the output and employment would suffer further due to price rise for an economy operating well below full employment and full capacity level of activity.
Compliance under the GST Regime
It has also been claimed that there would be better compliance under the GST regime, at least in the long run. Compliance is defined as the ratio of the actual tax collection to potential tax collection, given the tax rate and tax base. This compliance is typically dependent on the efficiency of tax administration, the tax information network and on the degree of corruption. The present situation of information network is such that two or three main cities account for almost the entire VAT collection of the states (Dasgupta 2016). The retail business is largely informal and the government does not have proper information about sales in this sector.
During the introduction of VAT, we were told by government economists that leakages would automatically minimise because of input tax credit system, since there would be incentive to collect invoice/cash–memos, and that compliance would improve for sure. It is debatable whether the VAT collection has somewhat increased because of the high growth in the base, or because of better compliance.
I would like to argue that the introduction of GST is neither a necessary nor a sufficient condition for improvement in the tax compliance. If the degree of corruption decreases, the efficiency of tax administration improves and the tax information network expands, the compliance may improve even without the introduction of GST. If there is no improvement in these factors , the compliance may even deteriorate after the introduction of GST.
Some argue that the central tax administration is more efficient and less corrupt than the state government machinery, and hence the GST collected by the joint authority would necessarily improve compliance. However, this is a baseless assumption without any empirical evidence to substantiate it.
Therefore, there is absolutely no reason to believe that tax compliance would automatically improve because of the introduction of GST per se. In fact, if the tax rate is too high, there would be a higher tendency to evade it. And if multiple taxes are clubbed into one tax, one evasion would be equivalent to multiple evasions. The perverse incentive for a parallel economy to grow would also be higher. How practical would it be to increase the tax rate on high-valued precious metals like gold from 1% to 6%, suddenly at one go, is also an important question. There is a possibility of an increase in various kinds of unlawful rent-seeking activities and so on. Therefore, whether there would be a net gain or loss in terms of compliance in the days to come, is ambiguous.
Controversy Regarding Revenue Neutral Rates
Clearly, the RNR is one of the most important considerations as no government would like to face losses in revenue due to a tax reform. Whether a particular state government would incur revenue loss or not, would be determined by the difference between the actual SGST rate (plus the effective rate of transfer) and the RNR of that particular state. Understandably, the RNR would be higher for the relatively more industrialised states with a comparatively larger base for excise duty collection. However, there is no consensus regarding the calculation of RNRs, which is of utmost importance. Bhaskar and Nath (2015) mention that the report of the National Institute of Public Finance and Policy (NIPFP) has preliminarily determined the RNR as 27%.
Following the format in the Report of the 13th Finance Commission, Kavita Rao (2010) opined that if India is at the lower end, the base for the tax would be about ₹15 lakh crore and not ₹21 lakh crore as cited, and the corresponding RNR will be more than 23%. The Ministry of Finance (2015d) says that the RNR for the centre and the state is estimated to be 5.64% and 6.34% respectively. The combined RNR is determined at 11.98%. The latest estimate of combined GST by the committee of the Ministry of Finance headed by the chief economic advisor (CEA) is 19%, and that by Kavita Rao of the NIPFP is around 23% based on different assumptions (Kavita Rao 2016). Therefore, there is a huge confusion regarding the RNR. There is no question of the average combined RNR being less than 20%. If the actual GST rate is less than the RNR, the combined indirect tax revenue of the centre and the states would come down as a proportion of GDP and if the actual rate is higher than the RNR, there would be revenue gain but it would be, ceteris paribus, inflationary and detrimental to inequality and growth.
Conclusions and Policy Implications
The question to be confronted here is not of the feasibility of amending the constitution for implementing GST; the vital concern is about the future of Indian fiscal federalism, and the degree of inequality within our democracy. Hence, cost-benefit analysis must be done more seriously before undertaking such a big tax reform in the country. The GST should be designed more carefully from a more holistic point of view considering the Indian context, keeping all the possible trade–offs in mind. The policy recommendations that logically emerge from the above discussion can be articulated in the following manner.
More emphasis should be put on direct taxes for improving India’s abysmally low tax–GDP ratio for arresting the inequality situation in the country from worsening further.[ii] Each individual state should have the freedom to set their own SGST rates for different commodities and services with some provisions for entry tax in order to preserve the fiscal autonomy of the state governments[iii]. The calculation of RNR for the central government should be done separately for determining the CGST rates, which would be uniform across all the states. Since the centre has to give some compensation to the states with larger excise duty bases, and since the central revenue receipts from GST would be part of the divisible pool, the weighted average CGST rate (as a percentage of GDP) should be made accordingly so that the central government also does not make any net revenue loss. The IGST rates would naturally vary from state to state depending on the different SGST rates of respective states. Also, the government should take the required extra steps for enhancing growth and keeping inflation in check as implementation of the GST, per se, would not necessarily help in achieving these objectives. The government should also make conscious efforts to improve tax-compliance as the introduction of GST cannot automatically ensure it. If compliance improves, obviously, the GST rate also does not have to be too high. As a result, the incentives of tax evasion are also likely to come down.
[i] From 2004–05 to 2013–14, Ministry of Finance (2015b) .
[ii] In this context, it is worth mentioning that practically there is no property inheritance tax and/or wealth tax in this country. We must also reduce the tax expenditure or revenue foregone as proportion of GDP. Some economists argue that the direction of expenditure policy should be made pro-poor but the tax policy should not be necessarily based on equity principle. But, if the proposed tax policy is surely inequality creating and demand contractionary, then, by this logic, the pro-poor expansionary expenditure policy should be made a precondition for its implementation! Some tax experts also point out that garnering more revenue from the direct taxes is difficult as the base for direct tax is very small in countries like India and, hence, we have to rely more on indirect taxes. Does that mean we should tax 80% of our poorer population more, whose family income is less than the threshold of direct tax exemption? (Das, 2014b).
[iii] Kavita Rao (2010) has also argued that there is a 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.
Bagchi, Amaresh (2007): “International Lessons in Goods and Services Tax,” Economic & Political Weekly, Vol 42, No.45–46, pp 28–32.
— (2006): “Towards GST: Choices and Trade-offs,” Economic & Political Weekly, Vol 41, No 14, pp 1314–1317.
Bhaskar, V and Kailash Nath (2015): “What Ails the Implementation of the Goods and Services Tax?,” Economic & Political Weekly, Vol 50, No 37, pp 12–14.
Das, S (2014a): “Crossing the Kautilya-Rekha of Taxation,” Vikalp, 26 February.
— (2014b): “On Neglected ‘Middle Class’ in India,” Vikalp, 3 November.
Dasgupta, Chirashree (2016): “The Flawed Premises of GST: Taking a Second Look at the Biggest Tax Reform in a Long Time,’ Governance Now, 20 August, http://www.governancenow.com/views/columns/the-flawed-premises-gst.
Kaldor, Nicholas (1959): “Tax Reform in India,” Economic & Political Weekly, Vol 11 , No 4–6, pp 195–198.
Kalecki, M (1943): “Political Aspects of Full Employment,” in Selected Essays on the Dynamics of the Capitalist Economy—1933–1970, Cambridge: Cambridge University Press.
Kumar, Arun (2015): “Macroeconomic Aspects of Goods and Services Tax,” Economic & Political Weekly, Vol 50, No 29, pp 26–30.
Ministry of Finance (2015): “Concept Note on GST,” Department of Revenue, Ministry of Finance, Government of India.
— (2015a): “Indian Public Finance Statistics 2014–15,” Ministry of Finance, Government of India.
Musgrave, R (1959): The Theory of Public Finance: A Study in Public Economy, McGraw-Hill Book Company, Inc. Asian Students’ Edition, New York, Toronto, London.
Patnaik, Prabhat (2016a): “A blow against federalism: The Implications of a Uniform Goods and Services Tax,” Telegraph, 28 June.
— (2016b): “The Debate on GST,” Peoples’ Democracy, Vol 40, No 38.
Rao, Govinda (2009): “Goods and Services Tax: Some Progress Towards Clarity,” Economic & Political Weekly, Vol 44, No 51, pp 8–11.
— (2011): “Goods and Services Tax: A Gorilla, Chimpanzee or a Genus Like ‘Primates’?” Economic & Political Weekly, Vol 44, No 7, pp 43–48.
Rao, Kavita (2010): “Goods and Services Tax: The 13th Finance Commission and the Way Forward,” Economic & Political Weekly, Vol 45, No 48, pp 71–77.
— (2016): “New Assumptions, New Estimates: Scrutinising a New Report on Revenue Neutral Rate,” Economic & Political Weekly, Vol 51, No 4, pp 63–66.
EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.