When Machines Decide You Are a Fraud: NGTP Tags, Money Laundering and the Silent War on Bona Fide GST Taxpayers
1. Introduction: Advanced missiles aimed at small taxpayers
In the early years of GST, the main fear was the complexity of returns and e‑way bills. Today, the bigger threat for many honest taxpayers is something more invisible: back‑end risk engines and labels like “NGTP – Non‑Genuine Taxable Person”.
Departments now heavily rely on analytics, AI models, e‑way bill/e‑invoice linkages and “non‑genuine” lists to decide who is a fraud, often before any serious on‑ground verification.
In parallel, the Government has brought GSTN under the Prevention of Money Laundering Act (PMLA) information‑sharing framework, enabling ED and FIU to use GST data in money‑laundering investigations.
On paper, these tools target large fake billing rackets. In practice, they often hit small businesses and genuine entrepreneurs, who suddenly find themselves branded as dealing with “non‑genuine”, treated almost like money launderers, and facing huge demands of tax, interest and penalty—sometimes double taxation—purely on the strength of system flags.
This article explains how NGTP flags and non‑genuine lists actually work, how they connect to money‑laundering narratives, where departmental behaviour becomes unlawful or unjust, and what taxpayers and officers should both do to restore balance.
2. What is NGTP and a “non‑genuine” list?
“NGTP” stands for Non‑Genuine Taxable Person. It is not a term defined in the GST Act or Rules, but it is widely used inside departmental systems and circulars.
State GST departments like Maharashtra periodically release lists of “non‑genuine taxpayers”, with GSTIN, trade name, registration and cancellation dates, and a Unique NGTP Code. These lists cover taxpayers whose registration certificates are cancelled ab‑initio, treating them as non‑existent or bogus from the beginning.
According to professional literature and State documents:
These lists are built from investigations, site visits and data analytics highlighting suspected fake billing, circular trading, or non‑existent businesses.
Once a GSTIN appears in the non‑genuine list, officers across the State treat its invoices as suspect, and ITC used by recipients is either auto‑blocked or targeted for denial.
In theory, NGTP status is only a risk tag and a trigger for investigation. In reality, it often becomes a stamp of guilt, which the department then uses to reverse ITC across the chain without fully examining each recipient’s facts.
3. How the analytics engine works (GSTR‑1/3B, e‑way bills, AI)
The current GST architecture is highly data‑rich. Return data, e‑way bills and e‑invoices flow into back‑end systems that run risk algorithms. Typical red flags include:
Return mismatches: Large outward supplies in GSTR‑1 but low or nil tax payment in GSTR‑3B; repeated non‑filing of returns.
Abnormal ITC utilisation: Very high ITC compared to liability, 90–100% utilisation for many months, or more ITC passed on than is consistent with the entity’s capital and turnover.
E‑way bill anomalies:
High‑value invoices without corresponding e‑way bills where they are mandatory.
Unrealistic distances or vehicle numbers (for example, the same truck moving goods in different states on the same day).
E‑invoice (IRN) and e‑way bill mismatch: For entities above the e‑invoice threshold, the system cross‑checks IRN data, e‑way bills and returns to identify unusual trade chains.
From these parameters, risk engines generate risk reports and push them to jurisdictional officers. The officer may then:
Mark a GSTIN as NGTP or “risky”.
Propose cancellation (often ab‑initio).
Circulate the GSTIN in internal non‑genuine lists and initiate notices to all recipients who have claimed ITC on that supplier’s invoices.
The danger is that this entire chain can run with minimal or very poor physical verification.
4. Two kinds of “bogus” entities – real fraud vs machine‑made fraud
We must distinguish between two very different categories.
(a) Real shell/fake entities created by fraudsters
There is no doubt that large fake invoice rackets exist. TaxGuru and other analyses show how fraudsters misuse Aadhaar/PAN or use frontmen to obtain GSTINs with no real business, then issue huge volumes of invoices to pass ITC without any supply of goods or services.
Registration lapses—weak KYC, “deemed approval” without proper checks, mechanical Aadhaar verification, and superficial or no Rule 25 physical verification—allowed many such shell entities to enter the GST ecosystem.
Later, when these racks are detected, the department cancels such registrations, often ab‑initio, and brands them non‑genuine. ITC taken by buyers down the chain is then denied with demands, interest and penalty.
Here the basic fraud is real, but the policy question is: should bona fide buyers who had goods, invoices and bank‑channel payments be punished for the department’s registration failures?
(b) Machine‑made “bogus” entities – false positives
The second, more worrying category: genuine taxpayers who get wrongly classified as non‑genuine because of compliance lapses and analytics.
Examples from practice and commentary:
A small manufacturer with erratic filing and high ITC usage gets risk‑flagged; one casual visit when the premises is closed leads to a report of “no business”, and the GSTIN is branded bogus.
A trader shifts premises but does not promptly update registration; the inspector visits the old address, finds it closed, and records “non‑existent”, triggering NGTP tagging.
An SME facing cash flow issues delays 3B payments and utilises ITC heavily for a few months; the algorithm assumes “fake billing” because the pattern resembles a racket.
In such cases, the taxpayer may be non‑compliant but not fraudulent. Yet the system starts treating them like fraudsters, and their buyers get notices for “non‑genuine purchases”.
5. A simple example: how a genuine buyer becomes a money‑launderer on paper
Consider a small engineering unit in Karnataka—call it ABC Components—which buys steel from XYZ Steels.
System flags XYZ Steels
GSTR‑1: ₹50 crore outward supply.
GSTR‑3B: tax declared on only ₹3 crore.
ITC utilisation: 98% for several months, almost no cash.
E‑way bills: far fewer than expected, plus some anomalies in vehicle numbers.
The back‑end marks XYZ Steels as “high risk”.
Minimal physical work, quick cancellation
The officer receives a risk report, visits once; premises are locked.
Returned notices, no response, perhaps a wrong phone number.
Officer writes: “Non‑existent; no business; bogus”.
Registration of XYZ is cancelled ab‑initio and it is assigned an NGTP code and placed in the non‑genuine list.
Non‑genuine list travels, buyers get notices
The NGTP list is circulated internally; ITC on all XYZ invoices is now suspect.
ABC Components has purchased ₹2 crore of steel from XYZ, with tax paid through bank transfers.
ABC receives a notice:
“You have availed ITC of ₹36 lakh on purchases from XYZ Steels, GSTIN… which has been declared non‑genuine. Explain why this ITC should not be disallowed as bogus purchases.”
How the money‑laundering narrative enters
ED or FIU picks up the fake billing pattern of XYZ from GSTN data (now notified under PMLA for information sharing).
The “proceeds of crime” are assumed to be the ITC passed through XYZ’s paper invoices.
Anyone receiving ITC from XYZ, including ABC Components, is at risk of being painted as part of a laundering chain, especially if large volumes are involved.
Reality on the ground
ABC Components has a factory, stock records, goods receipt notes, and it has used steel in its own taxable supplies.
It has paid XYZ through bank, at market price, with GST.
It has no connection with any sham bank accounts or round‑tripping.
But in the departmental narrative, ABC is suddenly “dealing with a non‑genuine taxable person” and “benefitting from a fake billing/money‑laundering racket”. The demand raised is often tax + interest + 100% or more penalty, effectively collecting tax twice on the same goods—once from ABC, and again if anything is recovered from XYZ.
6. What is money laundering and how is GST now linked?
Basic concept of money laundering
Money laundering means processing the proceeds of crime so that they appear to come from a legitimate source. It generally involves three stages: placement, layering and integration.
Under PMLA, “proceeds of crime” are property derived from a scheduled offence. GST‑related fraud can become a predicate offence when it overlaps with offences like forgery, cheating, or other scheduled crimes.
How GSTN is pulled into PMLA
The Finance Ministry has recently included GSTN in the PMLA framework for information sharing. This means:
ED and FIU can share data with GSTN where GST‑related violations have a money‑laundering angle.
GST analytics can, in turn, support identification of laundering networks using fake billing to move and legitimise funds.
The Revenue Secretary has clarified that GSTN’s inclusion is for information sharing and FATF compliance, and is technically under section 66 of PMLA, not under GST law itself.
However, in practice, the narrative easily spills over: a large fake invoice network is presented as a laundering route, and recipients down the chain can be painted as part of the laundering ecosystem, even when they are small, bona fide businesses.
Can a small taxpayer be a money‑launderer?
Legally, yes—if a small taxpayer knowingly participates in creating bogus invoices, rotating funds, or helping conceal proceeds of crime, size doesn’t matter.
But most MSMEs and small entrepreneurs targeted today are not doing sophisticated layering. They are:
Buying goods they need;
Paying through banks;
Filing returns, sometimes late or imperfectly;
Struggling with compliance and documentation.
Treating such people as potential money launderers solely because they purchased from an NGTP is an overreach and a misuse of both GST analytics and PMLA logic.
7. Judicial push‑back: need for proper verification
Courts are beginning to push back against this mechanical approach.
In a recent Karnataka High Court case on “non‑genuine purchases”, the Court held that serious allegations must be backed by proper verification, and remanded the case when the department ignored additional evidence like e‑way bills, vehicle details and transaction records produced by the taxpayer.
Tax Guru’s recent pieces on retrospective cancellation and ITC denial also emphasise that:
Fake registrations reflect lapses of departmental verification and risk systems.
Yet, instead of owning these lapses, officers cancel registrations retrospectively and deny ITC to buyers, raising huge demands with interest and penalty.
Genuine assessees are left fighting litigation for years, while there is no visible accountability for officials who allowed those registrations in the first place.
This judicial and professional trend supports the argument that analytics and NGTP lists can only be starting points, not conclusive proof of fake purchases or money laundering.
8. How taxpayers and professionals should respond
For recipients (bona fide buyers) facing NGTP / non‑genuine notices, the defence must systematically attack the assumptions behind the analytics.
Step 1: Demand the underlying material
In replies and cross‑examinations, insist on:
The specific reasons and risk parameters for tagging the supplier as NGTP, not just the label.
Period‑wise reconciliation showing which invoices and which months have GSTR‑1 vs 3B vs 2B mismatches.
Detailed e‑way bill analysis: invoice‑wise, date‑wise and vehicle‑wise list where they allege non‑movement.
Copies of physical verification reports, photographs, statements, if any, used to declare the supplier non‑existent.
Without this, the taxpayer is fighting a ghost narrative.
Step 2: Produce your own factual trail
For your client’s purchases, put forward:
Tax invoices, purchase orders, goods receipt notes and stock records.
E‑way bills (where applicable), transport documents, weighbridge slips and vehicle details.
Bank statements showing full payment—including GST—to the supplier, with no cash‑back or circular flow.
This directly contradicts any broad allegation of “no movement of goods” or purely paper billing.
Step 3: Legally attack the over‑reliance on analytics
In your replies and written submissions, emphasise that:
Risk parameters and NGTP status are administrative tools for selection, not statutory tests of genuineness.
Courts (like Karnataka HC) require serious allegations to be tested against full factual material; mechanical branding as bogus due to return mismatch or one site visit is not enough.
Departmental failures in registration verification cannot be used to punish bona fide buyers who did reasonable due diligence.
9. A message to departmental officers: how to use analytics lawfully and fairly
This is not an argument against analytics or NGTP lists. They are necessary tools to fight large, organised fake billing rackets. The issue is how they are used.
A fair and lawful approach would include:
Treating system flags as starting points only. An NGTP tag should trigger thorough investigation of that supplier, not automatic denial of ITC to every recipient.
Ensuring proper physical verification under Rule 25: multiple visits if needed, speaking to neighbours, checking electricity and rental records, rather than recording “non‑existent” on one closed‑door visit.
Separating compliance defaulters from actual fraudsters: high ITC usage and delayed returns might be risky from revenue perspective, but they are not equal to participation in a fake billing racket.
Respecting judicial trends that favour bona fide buyers with documented receipt of goods and bank‑channel payments, even where a supplier later turns out to be problematic.
When officers use analytics as an advanced missile indiscriminately, they may show short‑term collection numbers, but they undermine trust in GST and generate large volumes of avoidable litigation.
10. Conclusion: Re‑balancing the system before trust collapses
India’s GST data system is powerful enough to detect complex frauds and support money‑laundering investigations. Bringing GSTN into the PMLA information‑sharing network strengthens that capability.
But when NGTP tags, non‑genuine lists and AI risk scores are used mechanically, without genuine verification and without distinguishing between fraud and error, the system turns its missiles on the very taxpayers it was supposed to support—small, bona fide entrepreneurs trying to comply.
The path forward is not to discard analytics, but to humanise its use:
For officers: treat flags as leads, not verdicts, and apply law and common sense before destroying someone’s business.
For taxpayers and professionals: understand how these tools work, demand full disclosure of the underlying material, and aggressively present factual and legal rebuttals when branded as dealing with “non‑genuine” or as part of a laundering chain.
If this balance is restored, the same digital tools that are now feared as advanced weapons can become what they were meant to be—precision instruments that target real fraud, while leaving honest taxpayers to do business in peace.
Author Bio
