With the compensation cess falling short, states may now have no option but to agree to raise GST rates at the lower end.
Despite his credentials as an economist and a former secretary general of Ficci, most tend to dismiss West Bengal finance minister Amit Mitra’s warning on the stagnant GST revenues as pure politics. At an Express Group event, Mitra spoke of India’s federal polity being at risk if the Centre didn’t extend the guarantee of a 14% growth in the states’ revenue from GST; right now, if the state GST revenues don’t rise by 14%, they are to be compensated for the shortfall. The states would fall off a cliff, Mitra said, if this guarantee wasn’t extended for another five years after 2022. While the situation may be more serious, in one critical aspect, than Mitra pointed out, the good news is that this may also result in much-needed GST reform.
Many reasons are given for GST revenues not being as buoyant as originally envisaged, ranging from the fact that important sectors have been kept outside the GST net to large-scale tax-theft, or even the fact that, as many say, the system was never robust enough to deliver… Whatever the reason, from Rs 92,581 crore in FY18, monthly GST revenues rose to just Rs 101,049 crore in April-September FY20 (see graphic).
As a result, compensation-cess collections have averaged around Rs 7,500-8,000 crore per month while, thanks to sluggish overall GST revenues—revenues need to grow at 14% every year for the states’ share to grow at the same rate—the amount that needed to be paid as compensation has skyrocketed; from Rs 6,000 crore per month in FY18 to Rs 11,500 crore so far in FY20. This means the compensation cess will fall short by November or December.
While Mitra and many others are hoping that the Centre will dip into its tax kitty to make good the shortfall, the GST law doesn’t say the compensation has to be paid by the Centre if cess collections fall short. So, even if the Centre guarantees a 14% growth in state GST revenues for another five years, in case the compensation cess falls short, the states get nothing.
In which case, while the states can blame the central government for not implementing a perfect GST—since GST is a joint responsibility of the GST Council in which all states are members, of course, a blame game may not even fly—this won’t really help. In all probability, the GST Council will have to agree to raise tax rates on items that are either taxed at very low rates or not taxed at all. The 5% tax slab, for instance, fetches the government Rs 120,000 crore (see graphic); doubling this rate will give a big boost to GST revenues and, at a time when inflation is so low, the consumer impact of a tax hike may not be as high as imagined. Similarly, the 12% bracket fetches Rs 70,000 crore, and an increase in this rate will add to GST revenues.
There are additional benefits of such rate-rationalisation. One, as GST rates are raised, and the gaps between tax slabs narrow, the incentive to cheat by using fake invoices falls. Two, as rates rise, the government will lose less from input tax credits. Mobile phones, for instance, are taxed at 12%, but their inputs at 18%, so the government ends up giving GST refunds to the industry. In textiles, the fact that the final cloth is taxed at 5% while inputs are taxed at 18% means an input-credit loss of Rs 30,000 crore a year; this loss is Rs 7,000 crore in the case of fertilisers where the final product has a 5% tax rate versus 18% for the inputs. Once GST rates are raised for the final product, such input-credit losses will reduce.
Faced with the need to hike rates on items of so-called mass consumption, most states will argue a rate-hike wouldn’t be needed if tax theft was less rampant. Junior finance minister Anurag Thakur spoke of a Rs 45,600 crore tax loss due to fraud, but this was the figure for all indirect taxes; the government caught input tax fraud of around Rs 11,500 crore last year, so it is possible GST evasion levels could even be over Rs 100,000 crore a year.
While a part of this will get fixed once rates are rationalised, a basic level of invoice-matching—the heart of the GST system—takes place even today. Everyone has to file their invoice-level returns as part of GSTR-1, so when firms pay their GST dues and file the GSTR-3B summary return while doing so, a basic check takes place automatically. If firm A gives invoice-level details of Rs 100 crore of supplies to firm X, and firm B shows it is supplying Rs 350 crore to firm X, but X shows a turnover of only Rs 300 crore, this problem gets caught even now. A more detailed level of automatic checks is being designed as the earlier one didn’t work. The new system, where all invoices are uploaded without filing an actual return, is supposed to take off by next April; a huge delay from the original schedule.
And, once every business files its detailed annual returns, this will also help catch tax-theft since all supplier invoices have to be accounted for, but this requires considerable speeding up. Annual returns for FY18 were to be filed by December 2018, but this has been extended to November 2019; no dates have been fixed so far for FY19 and FY20; similarly, it is to be hoped that the new system’s invoice-matching will be more rigorous to catch tax theft. But, catching tax evasion takes time, and the government can’t wait for this to succeed if it needs higher revenues fast; catching evaders and tax-rate hikes will need to happen side by side, and both will reinforce one another.