States' GST compensation may need to be extended beyond 2022
November, 27th 2019
States will need to be compensated for their revenue shortfall under goods and services tax (GST) even after 2022 — the sunset year for compensation under the law — because of slow revenue growth, a report commissioned by the 15th Finance Commission (FC) has noted.
It shows that states would require compensation of at least Rs 1.67 trillion in 2024-25, because none of them would be able to achieve 14 per cent growth every year.
While Karnataka would need the highest compensation among states, Delhi too would face a revenue shortfall to a great extent in coming years, the report warns.
Compensation cess is collected over and above the prevailing GST rate on certain luxurious items of consumption. The 15th FC might consider this an important input while it revises the formula for devolving tax revenue from the Centre to the states.
This finding of the report might put weight on the states’ side in the devolution formula. States are fearing a reduction in their share after the terms of reference of the FC are revised to include special provisions for defence spending.
This fiscal year (2019-20) states are facing an acute crisis due to a shortfall in revenue from all sources, including state GST (SGST) and sales tax.
They allege due to structural issues in GST, they would be left with no support after the compensation period ends in 2022.
On similar lines, 15th FC Chairman N K Singh has been calling for a simplification in GST.
chart “The issue of GST rates, exemptions, changes, and implementation of the indirect taxes is entirely within the domain of the GST Council. This leads to unsettled questions on the ways to monitor, scrutinise, and optimise revenue outcomes… Since both the FC and the GST Council are constitutional bodies, coordination between the two is now an inescapable necessity,” he said in lecture on November 22.
The report, authored by Sachhidananda Mukherjee and R Kavita Rao of the National Institute of Public Finance and Policy, a New Delhi-based centrally funded think tank, also highlights inconsistencies in the data provided by various sources on GST revenue.
“Unless the information from different sources (GSTR-1 and GSTR-3B) converges, any projection made from them may be erroneous,” the report notes.
GSTR-1 is the tax return form that includes supplies from a company (goods or service) while GSTR-3B is the tax return form that includes the tax paid, input-tax credit utilised, and the summary of purchases and output and turnover of the company for the pertaining period.
The report says due to different sets of compliances for the two return formats, the aggregate data populated from these two returns does not match, and makes projections difficult.
There is a 20 per cent difference between the compliances of these two monthly returns, and this only adds to the inconsistency, the authors said.
These projections form an extremely important input for the Finance Commission because it bases its recommendations on these.
The compliance for the summary monthly return (3B) has been struggling to cross 65 per cent even after more than two years of implementation.
The report also shows the tax liability does not match the sum of the tax paid in cash and the credit utilised (which should usually be the case) in some quarters, especially the early quarters.
In addition, the aggregate credit utilisation data populated from GSTR-3B does not match category-wise credit utilisation.
The report uses data from the revenue department in the finance ministry and the data thrown in by the GST Network (GSTN), the IT services back-end of the GST tax system.
The two throw different sets of data, making the analysis of GST unviable.
A robust IT system and better tax compliance would be the only way to tackle this mismatch, and better projections can be made only when the GST system stabilises, the report notes.
GST revenue has grown only 5 per cent this fiscal year, and the compensation paid to states has run into Rs 66,000 crore in the first six months, the data shows.