Harshad Mehta and the 1992 Indian Securities Scam

India's largest bank could not find its own government bonds. In the second week of April 1992 the investment department of the State Bank of India failed to reconcile its holdings against the Reserve Bank's Public Debt Office records, and the gap pointed at a single client: a Bombay broker named Harshad Shantilal Mehta, who had taken roughly Rs 500 crore out of the bank against receipts for securities nobody could produce.

A crore is ten million rupees. At the exchange rate of mid-1992, near Rs 30 to the dollar, Rs 500 crore came to about $165 million — enough, in the India of that year, to matter to the finance ministry. The Reserve Bank's own examination of the same books eventually put the deficiency in the bank's investment portfolio at around Rs 649 crore, of which some Rs 574 crore covered transactions with Mehta for which the State Bank had received neither the securities nor so much as a receipt acknowledging them. Under pressure Mehta paid roughly Rs 620 crore back between 13 and 24 April.

On 22 April a contact telephoned the news desk of The Times of India in Bombay to say that the country's largest bank had summoned its most famous broker and demanded that he square up. The reporter who took the story was Sucheta Dalal. It ran on 23 April 1992, and the bull market it ended had been the largest in Indian history.

Why Banks Had to Own Bonds

Indian banking in 1991 was arranged around a single number. The statutory liquidity ratio required banks to keep a proportion of their deposits in approved securities, overwhelmingly government paper, and by the start of the decade that proportion had climbed to 38.5 per cent. Add the cash reserve ratio held with the Reserve Bank, and more than half of every rupee deposited in an Indian bank was pre-committed by regulation. The Narasimham Committee of 1991 argued that a requirement on this scale had stopped functioning as a prudential rule and had become a device for financing the government at below-market rates, crowding private borrowers out of the credit system, and it recommended cutting the ratio to 25 per cent over five years.

Two features of how compliance was measured turned that obligation into a market. Banks reported their positions weekly rather than daily, and the test applied to an average across the reporting period rather than to each day's closing balance. A bank could therefore run below its requirement early in the week and buy government paper back before the Friday return, and a bank with surplus securities could lend them out in the interval. Whoever sat between those two banks, matching the one that needed bonds on Friday against the one that needed cash on Tuesday, had a business.

Mehta had that business. He was born in 1954 into a Gujarati family of modest means, spent part of his childhood in Raipur, came to Bombay, and worked his way up from a clerical job at New India Assurance through a series of broking firms before setting up on his own account. By 1991 his firm, GrowMore Research and Asset Management, was among the busiest intermediaries in the money market, and the financial press had given him a name that stuck: the Big Bull.

Ready Forward

At the centre of the scam sat the ready-forward deal, a repurchase agreement in all but name. One bank sold government securities to another for cash and simultaneously agreed to buy them back at a set price on a set date; the difference was interest, and the securities were collateral. Nothing about the structure was improper. Ready-forward transactions were the ordinary plumbing of the Indian money market, and the Reserve Bank permitted them in a defined list of securities.

Trouble lay in settlement. Ownership of government stock was recorded at the Reserve Bank's Public Debt Office in subsidiary general ledger accounts, and a transfer required an SGL form lodged with the PDO — a paper instruction, processed by hand, against ledgers kept by hand. Volumes in the late 1980s had outgrown that machinery, and the market's workaround was the bank receipt. Rather than move the securities, the selling bank issued the buyer a BR: a document confirming that it held stock of a stated description on the buyer's behalf and would deliver on demand. Jayanth Varma's contemporary account of the affair identifies this substitution as the structural hinge of everything that followed, because a bank receipt was a promise with no central register behind it and no independent evidence that the underlying securities existed (Varma, 1993).

A second convention compounded it. Although the transactions were interbank, they were arranged through brokers, and by the early 1990s cheques were routinely drawn in favour of the broker rather than the counterparty bank. Money that was supposed to move from one bank's account to another's passed instead through the broker's account, where it could stay for a few days, or considerably longer, before continuing to its nominal destination.

The Receipt for Nothing

Given a document whose only backing was the issuer's reputation, the scheme reduced to finding an issuer willing to write one against nothing at all. Mehta found two: the Bank of Karad, a small institution in Maharashtra, and the Metropolitan Co-operative Bank in Bombay. Both issued bank receipts for government securities they did not hold and in some cases had never held, and both were wound up afterwards.

Those receipts then went to large banks as collateral for ready-forward borrowings. A public sector bank, or a foreign bank running an Indian treasury book, handed over cash against a BR it did not verify because BRs from counterparty banks were not, as a matter of practice, verified. The cash reached Mehta's account, and from there it reached the Bombay Stock Exchange. Nine or more banks were eventually named as having issued or accepted irregular receipts, the State Bank of India and the National Housing Bank among them.

The National Housing Bank case was the one that damaged the Reserve Bank most directly, because the NHB was its own wholly owned subsidiary. Securities transactions of roughly Rs 1,200 crore ran through the NHB on Mehta's behalf, against a portfolio that could not support them, under a chairman — Manohar Pherwani — who had previously run the Unit Trust of India and was among the most senior figures in Indian finance. Standard Chartered Bank, dealing through a different broker, Hiten Dalal, went to the Central Bureau of Investigation in June 1992 alleging that it had been defrauded of Rs 1,239 crore, close to $400 million.

The Big Bull

What the money bought was a narrative. Mehta argued that Indian equities had been mispriced for a generation because the market valued companies on historic book figures rather than on what it would cost to rebuild them — the replacement cost theory, as it was called in the Bombay press through 1991 and 1992. Applied to cement, the argument produced a target price for Associated Cement Companies far above anything the share had traded at, and ACC became the emblem of the boom. The stock went from around Rs 200 to nearly Rs 9,000 in a matter of months.

Bombay's benchmark followed. The Sensex closed 1990 at 1,048 and 1991 at 1,909, an advance of 82 per cent in a year in which the government had pledged gold abroad to meet its external obligations. From about 1,194 in April 1991 the index reached 4,467 in April 1992, a gain of 274 per cent over twelve months, and then gave back more than two fifths of it. By August 1992 it was near 2,529.

The December values are year-end closes. The April 1991 figure is the base from which the final year of the boom is usually measured, and the April 1992 figure is the high print of the boom's last week, reached immediately before Dalal's report appeared on the 23rd; the August value is a month-end level rather than the exact trough.

Mehta spent accordingly, and publicly. He kept a sea-facing apartment in Worli of some 15,000 square feet with a private golf patch, imported a Toyota Lexus when such a car was a novelty on Indian roads, posed for magazine photographers among the fleet, and paid income tax in amounts large enough to be reported as news. A broker who had been unknown outside the money market five years earlier was now the most recognisable businessman in the country, and small investors followed his positions on the assumption that he knew something they did not. He did. He knew where the money was coming from.

Janakiraman

To establish what had happened, the Reserve Bank appointed a committee under its deputy governor, R. Janakiraman. It delivered an interim report on 1 June 1992 and a second on 6 July, and in a sequence of further reports put the scale of funds improperly diverted from the banking system at about Rs 4,024 crore — roughly $1.3 billion at the rates of the day (Janakiraman, 1992). A Joint Parliamentary Committee took the inquiry further and reported in December 1993. Its finding on the banks was blunter than anything the Reserve Bank had written: "In many banks like the National Housing Bank, controls did not exist. In others, such as the State Bank of India, they existed but broke down partially or wholly because of the negligence of one or more of the functionaries."

Prosecution needed machinery that did not exist. Parliament passed the Special Court (Trial of Offences Relating to Transactions in Securities) Act in 1992, which attached the assets of the notified parties, appointed a Custodian to hold and realise them, and created a dedicated court in Bombay to try both the criminal charges and the civil claims. Mehta was arrested by the CBI on 9 November 1992. He faced 72 criminal charges and more than 600 civil suits.

The Suitcase

On 16 June 1993, with Ram Jethmalani acting for him, Mehta held a press conference in Bombay and said that in November 1991 he had personally delivered Rs 1 crore in cash, in a suitcase, to the Prime Minister, P. V. Narasimha Rao, at his official residence. Rao's office did not dispute that a meeting had taken place. It denied that any money changed hands, and no charge was ever brought on the allegation.

The claim did what Mehta presumably intended, which was to convert a banking investigation into a political crisis; it also fixed his public character for good. Debashis Basu and Sucheta Dalal's book-length reconstruction, published the same year, treats the suitcase episode as characteristic rather than aberrant — a man who had spent two years persuading institutions to accept his word in place of collateral, attempting the same trick on the electorate (Basu and Dalal, 1993). Dalal, looking back three decades later, was terse about what reporting it had cost her: nothing, she said, comes without a fight.

What the Scam Built

Indian market infrastructure as it exists now was largely assembled in the five years after April 1992, and most of it was a direct answer to something Mehta had exploited. The Securities and Exchange Board of India, which had been created as an administrative body in 1988, received statutory powers under the SEBI Act of 1992. The Reserve Bank stopped broker-intermediated ready-forward deals in government securities and required banks to deal directly with one another, moved settlement onto a delivery-versus-payment basis so that cash and securities changed hands simultaneously, and phased out the physical bank receipt.

Bombay's exchange was replaced rather than reformed. The National Stock Exchange was incorporated in 1992, recognised in 1993 and began trading in 1994, with screen-based order matching and a clearing corporation standing between counterparties — an institutional answer to the broker-dominated open-outcry floor, and one that forced the Bombay exchange to follow. Dematerialisation came next under the Depositories Act of 1996, which took paper share certificates out of the settlement chain and with them the forged transfer forms that had let some 2.8 million shares in around ninety companies be misappropriated during the scam years. Ajay Shah and Susan Thomas, surveying the sequence afterwards, treat the 1992 scam as the event that made this programme politically possible, having demonstrated that the existing arrangements could not be patched (Shah and Thomas, 1997).

Nothing in the pattern Mehta worked was new, and the pieces of paper were the whole of it. Anthony De Angelis had discovered in the early 1960s that American banks would lend against warehouse receipts for soybean oil without inspecting the tanks, and the Salad Oil Swindle broke two Wall Street brokers when the tanks turned out to hold seawater. Kuwait's unofficial exchange was building a comparable structure on postdated cheques at the same moment, and the Souk al-Manakh collapsed in 1982 when the cheques came due. What distinguished 1992 was the counterparties: not credulous lenders on the fringe of a market but the central bank's own subsidiary, the country's largest commercial bank, and the Bombay treasury desks of international banks, all accepting receipts they had no procedure for checking.

Settlement failures of this kind were understood elsewhere. New York had spent the late 1960s buried in unreconciled stock certificates, and the Wall Street paperwork crisis closed the exchange one day a week until the back offices caught up. Three years after Mehta's arrest a single trader's unreconciled account at a Singapore subsidiary destroyed a 233-year-old British bank, and the collapse of Barings in 1995 turned on the same failure to match a position against a document.

Mehta died on 31 December 2001 in custody in Thane, aged forty-seven, after complaining of chest pain. Of the criminal charges that reached trial, four convictions were recorded against him; the remainder of the docket was never finished. Bank officials were still being sentenced for scam-era transactions in 2017 and 2019, a quarter of a century after the fact.

The Special Court created to try him outlived him by a wide margin, and the Custodian went on distributing what could be recovered to the banks the money had come from — matching claims against transactions, receipt by receipt, long after the receipts themselves had been abolished.