In the first term of the National Democratic Alliance (NDA) government, digitization was next only to its two big achievements, the Insolvency and Bankruptcy Code (IBC) and the goods and services tax (GST)—both deeply flawed, but reformed to some degree, and further reformable. Digitization has facilitated the lives of the poor who are dependent on remittances, and has improved the beneficiary-targeting of an assortment of government welfare schemes. We are talking about nearly 25 billion retail digital transactions (in number, not value) in 2018-19.

For those incentivized or forced to move to digital platforms, an internet shutdown is a severe breach of contract. It leads to cessation of e-commerce, including Aadhaar-enabled payments at small retail outlets, digital banking transactions, including all one-time-password-related services, and wage payments under the Mahatma Gandhi National Rural Employment Guarantee scheme. On top of it all, automated teller machines (ATMs) cannot dispense cash. Shutting down the internet comprehensively obstructs consumption at a time of a consumption-led slowdown. It is not a recipe for macroeconomic stability. It leads to cessation even of GST payments at a time when GST revenues are struggling to meet budget estimates.

On GST, the revenue decline is a predictable consequence of the consumption and growth slowdown. A number of steps are reported as under consideration, one of which threatens to add to the biggest problem with GST so far—constant rate tinkering. Certainly, the present rate assignment across goods and services, more than the rate structure itself, is desperately in need of correction. But there is far greater scope for raising GST revenue by improving compliance, and in a way that promotes rather than hinders growth.

Two stroke-of-the-pen reforms are readily at hand. The first is to reintroduce the reverse charge mechanism (RCM), so that small suppliers are not discriminated against, as they have been so far by the GST structure. This will be immediately growth-friendly. The second is to link the e-way bill to GST returns, instead of just letting data on goods movements lie unused.

On GST rates, it is true that the average GST rate was too low to be revenue neutral, even at the time of its introduction. How low exactly? I did an exercise with the latest available 130-sector input-output table for 2013-14. I struggled through the initial rate list for goods and a separate rate list for services to map it on to 130 sectors, and found out first-hand the mindless complexity of the rate differentiation currently in place. That in itself is a compelling story, but will have to be told another time.

The weighted average GST rate worked out to 13.5%, at the relative sectoral shares of 2013-14. That the GDP was lower that year does not matter. My average does not include the cesses levied under the Compensation Cess Act to fund the 14% annual revenue increase guaranteed to states. Even those cesses are not yielding enough to honour that promise to states—such is the extent of the current slowdown. That guarantee should have been configured to a buoyancy, but thereby hangs another tale.

The 13.5% average at the initial rates of levy is much below the recommended rate of the Arvind Subramanian committee, which lay in the range 15-15.5 %. It does not adjust the taxable base, the denominator, for the commerce conducted by dealers below the taxable threshold, but neither did the committee. The weighted average would be lower today after the substantial rate reductions in September and November 2017 followed by later dribbles.

The current focus has to be on improving GST compliance. When compliance improves, revenue realized goes up even while leaving untouched the underlying prescribed structure of rates and exemptions. Compliance monitoring has to start with upstream producers of key intermediates like feedstock for polyester-based textiles, and other intermediates widely used like steel, aluminium, cement or rubber. This is where the e-way bill data will help. The problem, of course, is that two universal intermediates, petroleum and electricity, are outside the GST. It should nevertheless be possible to chase down the value chain of bulk buyers of petrol, diesel and electricity. The data will have uses other than revenue, such as assessing energy efficiency as part of the effort to mitigate climate change.

In conclusion, let me break down the average GST rate of 13.5% into its sectoral components. Food is taxed at 5.5%, agro-based manufactures at 10%, other manufactures at 17.8%, and services at 13%. On the face of it, these seem orderly, but there is a lot in the detail that could be cleaned up to yield higher revenue. Unjustified exemptions need to be removed, and with that, it will be possible to even out the high rate peaks. The comprehensive rate re-examination will take time and care. A rare piece of recent good news is that GST rate reform will be done only after the budget. While cleaning out exemptions, it must be remembered that vendors will be enabled to claim input credit, which was not possible when the product sold was exempted. The revenue gain has to be configured accordingly. A recent announcement that the GST council will not allow telecom companies to claim input tax credit against the additional revenue claim on them is alarming. Input tax credit is intrinsic to the GST, and cannot be denied in a capricious way.

*Indira Rajaraman is an economist