Exaggeration of evasion is dangerous

Being an ex-insider, I know for certain that evasion of tax is exaggerated by departmental officers. They get brownie points for showing more detection of evasion. Very few cases succeed. I am now writing only about GST. My immediate provocation of writing on this subject is that several reports are coming in newspapers about the detection of evasion.

A recently retired finance secretary wrote in his article in The Times of India on December 10, 2018, that for a revenue collection of ~1 trillion, ~4 trillion is settled by input credit, and therefore, if leakage of input credit of 10 per cent can be stopped, the collection will increase from ~1 trillion to ~1.4 trillion. This interesting arithmetic of 40 per cent evasion is a dangerous exaggeration. Adjustment of input credit is never 80 per cent, but a maximum of 50 per cent. The percentage was between 11.72 (1986) to 36.66 (1995) as found in a study published in Asia Pacific Tax Bulletin 5

(1996) at page

143. So the statement that if input credit misuse is stopped, revenue will increase by 40 per cent is a gross exaggeration.

The figures reported by the minister of state for finance in the Lok Sabha (reported on December 19, 2018, in The Hindu) show that between April to November 2018, 3,196 cases of evasion amounting to ~12,766 crore were detected. Now we must understand what these cases are. Mostly these cases arise due to mistakes in filling returns, non-filling of returns, mismatch of data and many such procedural mistakes, which are common due to newness of GST and taxpayers being not used to return forms. The computer-based matching of input tax credits has not been taking place due to IT-related issues. Therefore, credit is being allowed on a selfassessment basis as earlier. Even though invoice level details of outward supplies are required to be filed in GSTR1, a system-based matching is yet to be implemented. So now the old system is being followed. As we are aware, the return system is being redesigned. My visits to some countries in Europe, namely the United Kingdom and the Netherlands, and also Canada, some years ago gave me the knowledge that such invoice matching was not done. Those who tried it gave up. They try the method of intelligence-based detection. Systematic analysis of data of input and output is a time-tested method of finding out how much of evasion an industry can resort to. There are many industries that have fixed input-output correlation like motor vehicles, air conditioners, and refrigerators. There the number of outputs can be accurately verified and there is no scope for evasion.

The then minister of state on September 5, 2012, informed the Lok Sabha that nearly 18 per cent cases succeeded in tribunals, about 30 per cent in high courts and 9 per cent in the Supreme Court in the years from 2008. The latest data is not available. But that isn't likely to be much different.

It is well known that 3.67 per cent of registered taxpayers (which is less than 1,00,000) pay about 79.52 per cent of tax. It is easy to cover them by intensive audit. Another fact which is relevant is that nearly 20 per cent tax is paid by government undertakings. And there are many big and respectable private groups of companies like Tata, Aditya Birla, Mahindra and Mahindra, Hindustan Unilever, Sundaram-Clayton, Infosys, and Wipro that have never resorted to invoice manipulation. I know it for certain having been in the department for so long. So ultimately the evasion is much less than what is talked of nonchalantly by those who do not know.

The reason why such blatant exaggeration of evasion figures is dangerous is that it distorts the policy decision. If it is believed that there is too much of evasion, which is not real, then the department will bestow a lot of emphasis on invoice matching, which is just not practical and unnecessary, and will spoil the fluency and efficiency of the GST.

Conclusion: Evasion is much exaggerated. That is dangerous as it distorts the policy decision. Unduly harsh measures may be taken. Invoice matching is just unnecessary. Intelligencebased checking and better auditing are enough. I suggest that the National Institute of Public Finance and Policy be given a project to find out how much evasion is there.

The writer is member, Central Board of Excise & Customs (retired) Email: smukher2000@yahoo.com

EVASION ISMUCH EXAGGERATED. IT DISTORTS THE POLICY DECISION... INVOICE MATCHING IS JUST UNNECESSARY