GST features contributing to GDP growth slowdown
I was a fervent supporter of the idea of the goods and services tax (GST). I still am. But the manner in which it has been configured has thrown a satanic spanner into the Indian growth story. It is not just a matter of one-off pre-GST de-stocking. Were that the case, we would not need to worry at all.
The multiple rate structure is not the culprit for the growth slowdown. Yes, the many rates complicate life, especially for retail grocery stores hosting a wide range of products, but aside from making them groan, it is not really making them scale down their operations. Then there is the needless complexity of the GSTR forms with the superfluous information asked for, and the software requirements for uploading returns prepared offline. These are problematic especially in the hinterland, but have not by themselves led to shutdown or scale-down.
The principal culprit is the monthly frequency of reporting required under the GST (for businesses with annual turnover more than Rs75 lakh). Within each month, there are three dates in sequence for voucher uploading, consolidation and claims, with a daily penalty beyond deadlines crossed, added to interest on any tax credits denied. This formal voucher-based monthly reporting has dealt a death blow to the risk-sharing mechanism underpinning the efficiency of the Indian retail supply chain as we know it. And that is what has hit growth.
Take a retailer of non-perishable items like garments or footwear. Retailers order a consignment from upstream wholesalers according to their best judgement of what clients will buy. The traditional practice was that if a retailer overestimated the appeal of a new style to his client catchment area, he returned unsold stock to the wholesaler, and finally paid the wholesaler a few months later only for his net purchase, net of returned stock.
Risk cover does best when risk is pooled across many locations with diversified patterns of incidence. The wholesaler is able to bear the risk of sale reversal because he can re-distribute returned stock. A new style in garments or slippers may lie unsold in one location, but fly off the shelves in another. Wholesalers in turn spurn retailers who return stock beyond some percentage limit of the gross purchase, thus leaving enough risk with the retailer to incentivize him to judge his market correctly and put in his best sales effort. If goods are defective, the wholesaler in turn returns the stock to the manufacturer, which again assigns risk to the only level where defects can actually be addressed.
When there is a switch to monthly reporting, a wholesaler uploads the initial gross sale to each retailer, with GST charged on a numbered invoice lodged in the system. Although the GST system does permit reversal of sale through issue of a credit note which can be offset against the next sale to the same retailer, it adds to the procedural burden, and is not something wholesalers are willing to touch. In effect, sale reversal has become impossible under GST, even for defectives. Retail buyers are now being asked to take a consignment at their own risk, and thereafter hold their peace. The traditional risk-sharing mechanism lies shattered.
Given that the retailer can no longer (in effect) reverse any part of an uploaded transaction, he minimizes risk by reducing his gross purchase from the wholesaler to the floor of his expected range of retail sales. This is what has hit growth. Wholesalers faced with reduced retailer offtake in turn place lower orders from manufacturers. Manufacturers have responded by sharply lowering production, some operating at as little as 25% of capacity. There are circles of auxiliary services which have all been affected by this scale-down—unskilled labourers moving bales of cloth in wholesale markets, dyers, embroiderers, tailors. The damage to both production and employment is widespread and it is huge.
This is not a temporary supply shock, because it has made an enduring alteration to the business model of small and medium retailers. The state-level value added tax (VAT) by contrast required only quarterly reporting in some states. Others had monthly reporting, but there was always a flexible provision for netting out sale reversals. Quarterly reporting and/or flexibility with reversals was entirely compatible with the risk-sharing mechanism. Under GST, quarterly reporting survives only in the simplified (composition) scheme for those with annual turnover under Rs75 lakh, who pay GST on a specified percentage of their turnover with no voucher documentation.
Central excise did have monthly reporting with inflexible excise-bonded warehouses, but that was at the manufacturing level. Wholesalers have always assumed the full commercial risk of whatever stocks they contracted to take (minus defectives returned to the manufacturer who absorbed the full associated loss).
A major reason for monthly reporting under the GST was the matching requirement between vouchers uploaded by buyer and seller in business to business (B2B) transactions. Voucher matching is not a standard feature of value added taxation systems anywhere in the world, except in China’s Golden Tax Project (International Monetary Fund Working Paper 16/68). Were we doing China-chasing?
Voucher matching is quite distinct from the requirement of voucher uploading for sales and tax credit claims. Matching as a pre-condition for granting tax credits needlessly burdens the IT platform. If a legitimate credit is denied because of matching failure, the promise of the GST is belied. Think about it—GST is supposed to carry an embedded incentive for sellers to report all sales and GST collected, against which to set off their tax credit claims. If so, why not just let the incentive work, and confine checks to a fractional random audit of vouchers uploaded? Policing of every single B2B transaction (even through electronic matching) makes no sense at all.
The state VAT on goods did not require voucher matching. Input credit claims were self-declared, and cross-checked only when there was prima facie evidence of fraud, supplemented by random auditing. The biggest source of VAT fraud was inter-state sales on which no VAT was leviable. Businesses misreported a within-state sale as a sale to another state, and claimed a total refund on input taxes. That was why under the VAT system, trucks crossing state borders had to be checked and stood in long lines awaiting consignment inspection to cross.
Today, with integrated goods and services tax (IGST) payable on cross-state transactions, the advantage from fraudulently claiming inter-state sales has been chased away. And indeed, the only transactions for which voucher uploading and matching make enormous sense are inter-state sales liable for levy of IGST. If voucher matching is restricted to IGST transactions, the compliance burden gets reduced to manageable proportions.
For all other transactions not involving IGST, I believe we can dispense with voucher matching. It is still not too late to do so, since voucher matching begins formally from 10 October. I believe we can even dispense with voucher uploading and settle for a consolidated self-declared statement of GST collected and paid, with just a listing of all counterparties in B2B transactions, along with dates and voucher numbers (essentially form 3B, in the present design). This information will be sufficient for back office cross-checks of cases randomly selected for audit.
To summarize, monthly reporting has had a negative supply-side impact on growth. The formal reporting mechanism has made transaction reversal impossible, and thus destroyed the risk-sharing mechanism underpinning the efficiency of the Indian supply chain. Together with universal voucher matching, it has made for high compliance cost.
Aside from compliance cost, there are other factors contributing to the post-GST price rise. Let us say a GST-registered trader uses the services of a small courier company which is in the composition scheme and charges GST, but does not have to issue any vouchers. GST on that input cannot therefore be claimed by the trader buying these courier services (but if voucher matching is eliminated, it can be claimed against the GST registration number of the service supplier). Aside from robbing GST of its key appeal, unclaimable tax credits add upward pressure to retail prices.
The Dussehra to Diwali festival season is a major season for new styles and models. It is a period when we particularly needed the risk-sharing mechanism in place. In the growth figures for Q2 and Q3, the place to look is gross value added, or GVA (which will clean out the indirect tax element), and in particular, core GVA (excluding agriculture, which may have tanked in kharif 2017 and could give a false sense of growth decline, and excluding government, which marches to its own drummer).
Preliminary evidence suggests that big retailers of durable consumer goods are doing well. But it is at the small retail end that jobs are generated, in a spatially dispersed manner all over the country. It will be sheer folly to overlook the impact of GST, as it is presently configured, on jobs and livelihoods.
The solution is starkly clear. Shift to quarterly reporting for all. If that change is made, it will be met with a cheer audible across the entire country. Retain voucher matching for IGST transactions alone, and get rid of voucher matching for all the rest. Even better, get away from voucher uploading altogether, and we will have unshackled growth. At the same time, the frequency of random audit will have to be increased. The data from the initial experiment with universal voucher uploading can be used to understand the structure of Indian business transactions, and develop auditing guidelines.
I hope these correctives will be considered at the meeting of the GST Council today.
Indira Rajaraman is an economist.
Comments are welcome at email@example.com