Economy

Gradual Tax Reform Would Have Been More Effective Than a Shock GST

While government is of the view that the nature of the tax reform required it to be a shock, implementing it in a phased manner could have helped avoid much of the disruptions being experienced.

A BJP supporter holds a placard during a rally to support implementation of the Goods and Services Tax. Credit: Reuters

A BJP supporter holds a placard during a rally to support implementation of the Goods and Services Tax. Credit: Reuters

Last week, the Indian government tinkered once again with the tax reform introduced on July 1. While it is to be appreciated that the government is responding to the difficulties being faced in the economy, could the basic reform itself have been substantially different in the first place? Now the political view in the government – and sometimes even the academic view – appears to be that since the newly-introduced Goods and Services Tax (GST) involves goods and services, the Centre and the states, buyers and sellers and one indirect tax vis-à-vis a multiplicity of indirect taxes, it could not have been anything but a big bang disruptive reform.

However, the reform could have been gradual.

Consider the following hypothetical administrative system in the absence of the GST. In other words, the old-style taxes are still in place. The government picks up, say, the cement industry. A cement manufacturer is chosen at random, and told to provide a list of those it has sold its cement to. Suppose that cement has been sold to 26 different dealers (who, in turn, sell to builders who use cement in actual construction). Let us label these 26 dealers as A, B, C, … Z. Next, the government picks a dealer at random. Suppose this is dealer A. Now, the government inspects the accounting records of the cement manufacturer to find out the quantity and price of cement sold to dealer A. The government also inspects the accounting records of dealer A to find out the quantity and price of cement it had purchased from the manufacturer. If the quantity and price in the two separate records are the same (and the taxes are computed in accordance with the accounting records), then it is clear that there has been no attempt at hiding anything in this context. However, if the records do not match, then either or both parties are possibly involved in some wrongdoing. This opens the door for further investigation. On completion of the investigation, appropriate penalties, if needed, can be imposed on the wrongdoer(s). It can also help, as Suranjali Tandon and Kavita Rao have shown in a recent research paper, to include stigma alongside monetary liabilities.

The above illustration is in the context of the cement industry and its dealers. But the methodology is more general. It can be used for any industry and its dealers. Indeed, it can be used for any business-to-business transaction. This may involve a manufacturer and a dealer. Or it may involve one dealer and another dealer. It may also involve one manufacturer and another manufacturer. The application of methodology can be on a large scale. However, as already mentioned, it can be done selectively on a random basis. The idea is to keep a check on the costs of administration. Furthermore, there can be a possible check of the government officials to keep the probability of collusion between government officials and taxpayers low. This can help keep a check on wrongdoing. It can also check the use of bogus firms in evasion of indirect taxes.


Also read: Narendra Modi’s ‘Good and Simple Tax’ Was Neither Good nor Simple


In the above model, the investigating agency is from the central government rather than from any of the states because of the possibility of the seller being from one state and the buyer from another. In such situations, it would be appropriate for the government to be involved in a (random) check. The investigating agency of the government can, in turn, inform the concerned tax department, which can be at the state or central level depending on whether the indirect tax evaded falls in the jurisdiction of the Centre or the states, given the multiplicity of the old tax structure. The concerned tax department can take the necessary action to recover the tax under the proposed plan.

Detecting tax evasion

Observe that in this alternative model, no GST is involved. Thus, GST is not necessary to keep a check on evasion of indirect taxes. This is not to say that GST should not be adopted, instead that it should be done separately after tax compliance has improved significantly. That way, there is hardly any disruption to economic activities because the public authorities are, under the proposed plan, attempting to accomplish one objective at a time.

In contrast, GST – as it has been introduced – is a big bang reform that is at once targeted at keeping a check on tax evasion and at making the indirect tax system more efficient. There exists an alternative plan that can address both concerns and yet avoids significant disruption. Broadly speaking, the changes in the tax regime could have been brought about in two different phases:

  • Improve tax compliance within the framework of the multiplicity of taxes, and the old tax rates (basically, without introducing any GST), and
  • Introduce GST after tax compliance has improved substantially.

The old tax system that prevailed till July 1 already included a value-added tax (VAT). So, this makes it easier to match the records of the buyers and the sellers. Even where VAT was absent, in any case the records of the buyers and sellers need to match. This enables the investigating agencies to detect evasion of taxes and impose appropriate penalties where required. Ex-ante, given that the investigating agencies may match the records of buyers and sellers, the tax compliance can improve.

The above is a simplified and stylised picture of detecting possible evasion of indirect taxes. There is, of course, a need to use expertise on various details; the expertise can be based on theory and evidence from all over the world. The objective is not to achieve perfect tax compliance in phase (a); it is instead to bring about a significant improvement.

A gradual approach

Indira Rajaraman has suggested a random check instead of a comprehensive check within the big bang reform as it stands in practice. This is a welcome suggestion. However, it can be used within phase (a) of the plan proposed in this article before GST is to be introduced under phase (b) of the proposed plan.

It is true that phase (a) would need changes in the structure of tax administration and in the budget for tax administration. It would also entail an increased manpower. It would require changes in legislation. It would also require (more) co-ordination between Centre and states. Last but not the least, it would need political support and the political will to improve tax compliance. But all this is required anyway even under the big bang reform. In fact, all such changes are required on a smaller scale initially under the proposed plan as only phase (a) needs to be implemented to begin with.

In fact, even within phase (a), the changes can be brought in, say, five different sub-phases. These can be sub-phases (a.1), (a.2), (a.3), (a.4) and (a.5). Suppose that the Indian economy can be meaningfully divided into five sectors (for illustration). Then sub-phase (a.1) can be introduced in the first year, sub-phase (a.2) can be introduced in the second year, and so on till sub-phase (a.5) is introduced in the fifth year. After five years of implementing phase (a), the government can proceed to bring about phase (b). Note that it is the changes in phase (b) which need to be brought about in one go but the changes in phase (a) can be further subdivided. This is because it is only under phase (b) that the GST will be introduced under the proposed plan.

Phase (a) is, in a sense, preparatory as it keeps bringing ‘firms’ in the informal sector under the tax net, and as it keeps making the firms in the formal sector more tax compliant. This happens first for firms related to sector one, then to firms related to sector two and so on till firms related to sector five are covered.

The advantage of a gradual approach is that it gives time to businesses, tax departments and professionals like chartered accountants and lawyers to learn and adapt. The word spreads from one firm to another on the need to change and on the modalities of change. This avoids disruption. The Indian economy has already gone through the adverse aftereffects of demonetisation on output and employment; a recent CMIE report by Mahesh Vyas shows that 1.5 million jobs were lost in the first four months of 2017. Even though it is hard to be sure that this is a direct consequence of demonetisation alone, it helps to bring about tax reforms in a phased manner to the extent possible.

There is a large literature on Transition and Economics. Though much of this is in the context of transition from a planned economy to a market economy, the broader lessons can be applied to the transition from a variety of indirect taxes to GST in India. In this context, this literature suggests that a gradual approach can be better than a shock therapy.

Gurbachan Singh is visiting faculty, Indian Statistical Institute (Delhi Centre) and Ashoka University.