Inside a ₹450 Crore Bank Loan Fraud: What the Santosh Overseas Case Reveals About Shell-Company Risk

The searches were carried out by the ED’s Lucknow Zonal Office, acting under the Prevention of Money Laundering Act (PMLA), 2002. The ED’s money-laundering investigation originates from a CBI case alleging that Santosh Overseas Ltd availed roughly ₹450 crore in loans from a consortium of banks. Investigators allege the funds were diverted and layered through a network of shell entities, accommodation-entry operators, and related companies, using fake invoices and circular financial transactions. The raids targeted premises linked to the company’s promoters, associated entities, and facilitators.

Under PMLA, the ED cannot open a money-laundering case on its own initiative. Its jurisdiction is derivative: it needs a predicate offence, an underlying scheduled crime, here the CBI’s bank-fraud FIR, to already exist before it can act. Once the ED registers its own internal case record, called an Enforcement Case Information Report (ECIR), that record becomes the basis for its subsequent steps: summons under Section 50, search operations, provisional attachment of property, and arrests. The July raids are that search-operation stage; nothing in the public record indicates the case has moved further than that as of this writing.

Six years before the July 2026 raids, the CBI had already opened a file on a company by the same name. On 2 July 2020, the CBI registered an FIR against Bulandshahr (Uttar Pradesh)-based Santosh Overseas Ltd and its director Sunil Mittal, alleging a fraud of ₹424.07 crore against a seven-bank consortium led by IDBI Bank, which also included Andhra Bank, Union Bank of India, Oriental Bank of Commerce, UCO Bank, and Canara Bank. The specific allegation was that the company availed packing credit (short-term export finance) based on a fake purchase order from an associate or group concern.

The FIR further alleged the company maintained current accounts with banks outside the lending consortium and routed sale proceeds through those non-consortium accounts without the consortium’s knowledge, alongside substantial dealings with unregistered related parties.

Corporate registry records support the identity link between the two names. MCA data lists Santosh Overseas Limited (CIN U52100DL2007PLC161735) as incorporated 9 April 2007 in Delhi’s Okhla Phase I, with Sunil Mittal appointed Managing Director on 1 April 2008, the same director named in the 2020 FIR. The same registry record shows the company’s status as “Dissolved Under Section 54,” with ₹649.78 crore in open bank charges still recorded against it.

The link between the first two rows above is strong circumstantial corroboration (same company name, same director, same fraud typology), not an officially confirmed connection: no 2026 source explicitly states that the ED probe traces back to this 2020 FIR.

Two terms carry most of the weight in cases like this one. An accommodation entry is a bogus book entry, created by a shell company or “entry operator,” that dresses up the true source of funds as a legitimate transaction (a fake loan, a share-capital infusion, or an invoice), typically to launder unaccounted money or, in a bank-fraud context, to build a paper trail that justifies where diverted loan money went. The term originated in litigation under Section 68 of the Income Tax Act, the tax-law provision used to challenge unexplained cash credits.

Hawala is the other piece of the vocabulary: an informal, unregulated value-transfer system that moves money outside formal banking channels entirely. Under PMLA, concealing or disguising the proceeds of crime is itself a punishable money-laundering offence regardless of the channel used to do it, and the ED has powers to investigate and attach assets connected to hawala-linked transactions.

In practice, the mechanics documented across Indian loan-fraud cases combine a small set of techniques. Fake or inflated invoices between related shell entities justify fund movement on paper that has no real underlying trade. “Round-tripping” sends diverted funds abroad disguised as payment for fictitious imports, then brings the money back disguised as foreign investment. Loan-back arrangements between related parties mimic a legitimate lending relationship, including sham interest payments. And in trade-based money laundering, a shell entity sits between two real trading parties and skims the difference between an inflated invoice and the true value of the goods.

There’s a wrinkle in how Indian law actually treats the phrase “shell company”: it does not define one. The Companies Act, 2013 never uses the term, and the Ministry of Corporate Affairs confirmed to the Rajya Sabha in 2021 that no Indian law defines it. Enforcement agencies work from a set of behavioural red flags instead. The MCA’s own list overlaps closely with the indicators RBI has told banks to watch for in loan accounts specifically:

The MCA column is drawn from its published shell-company indicators: between 2018 and 2021, applying indicators like these, the government identified 238,223 suspected shell companies and struck off more than 200,000 companies from the Register of Companies under Section 248. The RBI column comes from its 7 May 2015 circular, RBI/2014-15/590, introducing Early Warning Signals (EWS) and Red Flagged Accounts (RFA): a loan account is classified as an RFA once one or more EWS indicators raise suspicion of fraud, and that classification triggers a mandatory investigation by the lending bank’s own Fraud Monitoring Group.

These same red-flag indicators are also the practical checklist behind confirming a company’s beneficial ownership. See KYCKART’s glossary on beneficial ownership for more on this due-diligence step.

Cases like this one don’t happen in isolation. Per RBI’s annual report data, Indian banks reported 10,114 fraud cases worth ₹48,021 crore in FY2025-26, against 23,722 cases worth ₹32,803 crore in FY2024-25: fewer cases, a sharp jump in total value. Part of that FY2025-26 total, ₹30,199 crore across 314 cases, reflects older frauds reclassified and reported afresh following a Supreme Court judgment dated 27 March 2023 that required banks to re-examine certain cases.

Within that FY2025-26 total, loans (the “advances” category) accounted for 8,640 cases worth ₹40,774 crore, roughly 85% of the total value reported, with public sector banks carrying the large majority of both cases (5,418) and value (₹35,709 crore). For a wider look at bank fraud in India across all reported categories, not loan fraud specifically, see KYCKART’s broader piece on bank fraud.

Read together, the MCA’s red flags and RBI’s EWS indicators cover overlapping ground from two different regulators pursuing two different mandates: MCA’s list targets suspected shell companies generally, while RBI’s EWS/RFA list is specifically aimed at fund diversion in bank loan accounts. The overlap between the two lists is what makes this typology recognizable to a compliance reviewer even before any court finding exists. The alleged mechanics in the 2020 FIR (non-consortium bank accounts, related-party transactions, a fake purchase order underlying the original credit) map closely onto EWS indicators RBI published back in 2015: related-party transactions, funds routed to entities where the stated end-use doesn’t match actual use, and front companies funded on borrowed money. If the 2020-to-2026 link holds, it suggests a case matching these red-flag patterns can take years to move from initial suspicion to an eventual enforcement outcome.

The FY2025-26 RBI data adds a second layer to that same point. Loan fraud accounted for 8,640 of the 10,114 total fraud cases reported that year, about 85% of both case count and value, concentrated heavily in public sector banks, which carried 5,418 of those cases and ₹35,709 crore of the value. RBI’s EWS/RFA framework, introduced back in 2015, was already built around ongoing, indicator-based monitoring of loan accounts, a decade before this FY2025-26 concentration data existed. That the loan-fraud share remains this large a decade later suggests the framework’s ongoing-monitoring emphasis targets a persistent structural risk in how loans get disbursed and tracked. For a compliance or fraud-intelligence team, the practical implication is that shell-company red flags (shared addresses, no discernible business activity, circular transactions) are worth checking against a borrower and its related parties on an ongoing basis. The MCA’s own admission that no statutory definition exists means detection relies on pattern-matching across these indicators.