One Nation, One Tax, One Database

India's Goods and Services Tax (GST) should count as among the landmark economic policy reforms in the country over the last three decades. It has already been bearing fruits in terms of the large expansion in tax base and revenues.

The GST also provides an excellent opportunity to harness the power of data and create a robust decision-support system for both GST administration and economic policy making. This is also important since the low hanging fruits in terms of expanding the tax base may have been harvested, and further increases in GST revenues will be critically dependent on the quality of data analytics to arrest leakages. Further, the GST Network (GSTN) database can form the basis for the creation of a decision-support system to provide insights about local and national economic growth trends. 

The GST’s business intelligence (BI) is currently supplied through a diverse set of entities. The Central Board of Indirect Taxes and Customs (CBIC) has two entities – Directorate General of GST Intelligence (DGGI) and Directorate General of Analytics and Risk Management (DGARM) – and an Advanced Analytics in Indirect Taxation (ADVAIT) portal. The GSTN has the Business Intelligence and Fraud Analytics (BIFA) unit, the National Informatics Centre (NIC) has its GST Prime which is integrated with the e-way bills database, and each state government has its own data analytics units. This patchwork is a legacy of the evolution of the GST system and also arises from the confidential nature of the GST BI.

Further, most data access, especially for state governments, happen on a long-drawn and ad-hoc request mode through file sharing. While the central agencies have access to the entire GST database and some access to outside databases, the state governments rely on the outputs of BIFA and GST Prime. They have limited access to the transactions of even their own taxpayers outside the state. 

The net result is a BI system with multiple silos, differential access to data, limited standardisation of risk analytics, and sub-optimal harvesting of the power of data analytics. 

So what can be done?

Identification of the major sources of GST evasion like fake registrations, fake e-way bills, fake Input Tax Credit (ITC) claims, circular trading, etc., requires access to different data sources. It can also enhance the quality of data analysis on these violations, which maximises the likelihood of detecting evasions while also minimises the likelihood of harassment of honest taxpayers.

For example, an e-way bill generated for cross-border sales can be validated by checking with data from the NHAI toll gates on the passage of the vehicle concerned. Or, in cases of circular trading where the chain involves taxpayers in other states, detection of evasion requires access to GST databases of other states. Similarly, the verification of the rising trend of ITC claims on Inter-state GST (IGST) from the supplies effected from outside the state requires access to returns filed and taxes paid by the dealers outside the state. Or, taxpayers who transact in cash without invoices can be detected through their income tax filings and bank accounts information. 

In this context, a BI system that’s able to draw on the entire GST database, the Income Tax database, the MCA 21 of the Ministry of Corporate Affairs, Vahan database on vehicle registrations, national highway toll gates data, bank account statements etc. can be extremely powerful. 

The granular and high frequency GST data, when linked with other databases, can be extremely valuable in macroeconomic policy making. It can address the acute deficiency of good quality current data on economic indicators which has been a major handicap in economic policy making in India. For example, using the Harmonised System of Nomenclature (HSN) Code, we can get information on the sales trends for important goods and services like automobiles, fast moving consumer goods, healthcare services etc. The granularity of the data allows to construct indices which are credible and real-time proxies about the health of the local and national economies. The National Data and Analytics Platform (NDAP) could be entrusted such analysis.

How can this be achieved?

First, it’s important to be clear about what constitutes access. The GSTN Back Office (BO) Portal workflow should be linked with different databases through Application Programming Interfaces (APIs). This would institutionalize equal access to the databases for the central and state GST units. It would also ensure that access is restricted only to defined fields and for specific purposes defined in the BO Portal’s BI. This can overcome the risk of leakage and abuse of this data. This data access can be standardized by limiting access only with sufficiently strong reasons, clearly defined sharing protocols, high enough authorization levels, to the least required data fields, data use trails are digitally available etc. This would address concerns of privacy, data safety, and audit.

Given the scope of the work involved, one approach would be for the GSTN to co-ordinate with various Departments and entities of the central government, and link the BO Portal with their respective databases. Another approach would be to create a full-fledged information intermediation entity, with its own portal, to co-ordinate access to different kinds of databases and facilitate its access to central and state GST units. 

In a large country, there will be a varying pace of adoption of any innovation by different states. Accordingly, some states will pursue data analytics more actively. Further, since they administer half the taxpayers and have the largest field presence, state units have the best field intelligence to undertake robust data analytics. The approach proposed will allow states to develop their data analytics from the BO Portal and the different databases linked to it. Those data analytics filters found effective could gradually get scaled up across states. 

Finally, this will also make states, which were encouraged to dispense with their IT systems and migrate to the GST BO workflow, realise the promised benefits of being part of the GSTN. 

Now that the GSTN workflows have matured, it’s time to connect with other databases and enhance the effectiveness of GST’s BI. The GST achieved one-nation, one-tax. It’s now required to achieve a one-GST, one database.