From 1850 to Satyam: How Indian Company Law Was Written

A Privilege with Conditions

Of all the favours a legislature can grant a merchant, the strangest is permission to fail without paying his debts in full. Limited liability means that if a company collapses, its shareholders lose what they put in and nothing more; the creditors, who may have supplied goods on thirty days' credit, absorb the rest. Put that baldly and it sounds like a licence to cheat. The nineteenth-century legislators who granted it thought so too, which is why they attached a price. The price was disclosure.

The 1850 Act did not confer limited liability at all. Act XLIII allowed a joint stock company to register, to sue and be sued through a public officer, and to hold property in a manner the old law of partnership made awkward. Its members remained personally liable for the company's debts. What the statute added was a registrar and a set of particulars: names of shareholders, the amount of capital, the purpose of the undertaking. Registration was the bargain's first half, offered before anyone had decided what the second half should be.

The second half arrived seven years later. England had passed its Limited Liability Act in 1855 and consolidated the position in 1856. India followed with Act XIX of 1857, which permitted companies to register with limited liability but kept banking and insurance outside its scope, on the reasoning that institutions holding other people's deposits should not be allowed to shrug off their losses. That exclusion lasted until Act VII of 1860. Even then the word 'Limited' had to appear at the end of the company's name, painted on its premises and printed on its letters. The law did not trust strangers to find out for themselves. It made the company announce its own reduced exposure every time it asked for credit.

That suffix is still with us. Section 4(1)(a) of the Companies Act, 2013 requires a public company's name to end with 'Limited' and a private company's with 'Private Limited'. I have explained its purpose to clients more often than almost any other provision, usually in response to some version of a question a garment exporter from Kondhwa once asked me across this desk.

'Madam, limited means nobody can touch my house, no?'

It means a creditor cannot reach his house for the company's debts unless he has signed a personal guarantee, which, as he had for his overdraft, nearly every small-company promoter in Pune has done. The word protects him far less than he imagines. It protects his suppliers rather more, because it tells them in advance whom they are dealing with.

Borrowed Statutes

For the next half-century Indian company law moved in step with English law, usually a few years behind it. The Indian Companies Act of 1866 (Act X) reproduced much of the English Companies Act, 1862. A further consolidating Act followed in 1882. The Indian Companies Act, 1913 (Act VII of 1913) was modelled closely on the English Companies (Consolidation) Act, 1908, and it governed Indian companies for forty-three years, including the whole of the swadeshi enterprise movement, two world wars and Partition.

It would be easy, and partly true, to call this a colonial import fitted badly to local conditions. The 1913 Act said very little about the institution that actually ran Indian industry: the managing agency, a firm (often a family partnership) that controlled a group of companies under contract, drew commission on their sales and frequently answered to nobody but itself. English law had no equivalent to borrow from, so the Act was largely silent. The silence lasted until the Indian Companies (Amendment) Act, 1936, which placed the first statutory limits on managing agents' terms and dealings, after two decades of complaints from shareholders who had discovered that owning a company and controlling it were different things.

In September 1913, a few months after the Act received assent, the People's Bank of India in Lahore suspended payment. Over the following two years dozens of Indian banks failed alongside it.

The colonial statute had no answer to them. Bank collapses were handled as ordinary liquidations, depositors queued for whatever the official liquidator could recover, and a law written specifically for banking companies did not arrive until the Banking Companies Act, 1949, thirty-six years after Lahore.

Permission Required

Independence changed the purpose of company law. It did not happen overnight. In 1950 the government appointed a committee under C. H. Bhabha to review the 1913 Act, and its report of 1952 became the basis of the Companies Act, 1956, which came into force on 1 April 1956 with 658 sections. That length was itself a statement. The 1913 Act had largely assumed that shareholders could look after themselves once disclosure was made. The 1956 Act assumed they could not, and that the state should do much of the looking.

The surrounding policy explains why. The Industries (Development and Regulation) Act, 1951 already required a licence to establish or substantially expand an undertaking in a scheduled industry. The Industrial Policy Resolution of April 1956, adopted the same month the new Companies Act took effect, reserved whole sectors for the state. Capital was scarce, foreign exchange scarcer, and the planners in Delhi wanted to decide where both went. Company law became one of their instruments. A private enterprise could exist, but it would ask permission at most of the points where it might otherwise have used its own judgement.

The Act asked often. Section 269 required central government approval to appoint a managing or whole-time director in a public company above a certain size. Sections 198 and 309 capped total managerial remuneration at 11 per cent of net profits and made the government the arbiter of anything above the prescribed limits. Section 370, later replaced by Section 372A, restricted loans and investments between companies in the same group. These were provisions about control, not honesty, and the distinction matters for what came later. A director could comply with every one of them and still keep two sets of books.

Even so, the 1956 Act was rewritten in response to scandal from its first decade. The Dalmia–Jain group affairs, investigated by the Commission of Inquiry headed by Justice Vivian Bose, produced a report in 1962 describing the movement of funds between group companies, an insurer and a bank in terms that made the managing agency system look less like a business model than a conduit. The Companies (Amendment) Act, 1963 followed, enlarging the government's powers to investigate and, under Section 408, to appoint its own directors to a company being run oppressively. The managing agency itself was abolished with effect from 3 April 1970 by the Companies (Amendment) Act, 1969. An institution that the 1913 Act had barely noticed was ended by statute fifty-seven years later, on a fixed date.

I came into practice in 2002, when the 1956 Act still governed everything and most of the licensing that justified its approvals had been dismantled in 1991. The result was a statute built for one economy operating in another. Approvals the planners had once needed were still demanded by forms nobody had redesigned. Before MCA21 went live in 2006, a balance sheet went to the Registrar on paper as Form 23AC, with copies, and a clerk at the counter would occasionally return a filing because the stamp of the company's seal had smudged across a signature. The seal itself was not optional until the Companies (Amendment) Act, 2015.

What the 1956 Act never quite managed was to make a company's numbers trustworthy. It had amended its provisions on auditors, on loans to directors, on inter-corporate deposits, more than two dozen times by 2013. It had a Department of Company Affairs, a Company Law Board and an inspection power. It had, in short, every mechanism for supervising what a company was permitted to do, and comparatively few for checking whether what a company said about itself was true.

A retired chartered accountant from Sadashiv Peth, who had audited textile mills in the 1970s, put it to me once with some satisfaction. 'Those days, government was checking everything. Only nobody was checking the cash.'

Riding a Tiger

The letter was dated 7 January 2009, a Wednesday, and ran to about four pages. Ramalinga Raju, founder and chairman of Satyam Computer Services, then the fourth-largest information technology company in India, addressed it to his own board of directors and copied it to the Securities and Exchange Board of India and the stock exchanges. It is an odd document to reread. Parts of it read like a balance sheet, parts like a confession, and one sentence has become the only line from Indian corporate history that most people can quote.

The figures came first. As at 30 September 2008, Satyam's books showed cash and bank balances of ₹5,361 crore. Of that, Raju wrote, ₹5,040 crore did not exist. Accrued interest of ₹376 crore was fictitious. Debtors were overstated by ₹490 crore, and a liability of ₹1,230 crore, which he described as funds he had arranged personally, did not appear in the accounts at all. The quarter's operating margin, reported at 24 per cent of revenue, had in fact been about 3 per cent. The gap, he explained, had begun small years earlier and had grown each quarter because every inflated figure required a larger one to sustain it. Then the sentence: 'It was like riding a tiger, not knowing how to get off without being eaten.'

The share price fell by more than three-quarters that day. The government suspended the board on 9 January, using its powers under Sections 388B and 408 of the Companies Act, 1956 (the same Section 408 that the 1963 amendment had written after the Dalmia–Jain inquiry), and nominated its own directors. Raju was arrested the same evening. In April 2009 Tech Mahindra won the auction for the company. A special court in Hyderabad convicted Raju and nine others in April 2015.

Those are the facts most readers remember. What interests me professionally is the three weeks before the letter, because they show the board doing exactly what the 1956 Act asked of it, and that turned out to be nothing like enough.

On 16 December 2008 Satyam's board approved the acquisition of two companies controlled by Raju's family, Maytas Properties and Maytas Infrastructure, for about 1.6 billion US dollars. The board included distinguished independent directors, among them a Harvard Business School professor and a former cabinet secretary. Institutional investors revolted within hours, and the deal was withdrawn the next morning. With hindsight the purpose is fairly plain: a related-party acquisition would have turned cash that did not exist into land that did. In the following fortnight several independent directors resigned. None of them had known, and Raju's letter said so expressly. That was precisely the difficulty. A board that approves a ₹7,000 crore related-party transaction on the strength of cash balances nobody has independently confirmed is formally competent and practically blind.

The auditors faced a harder question. Price Waterhouse had signed Satyam's accounts for years. The fictitious balances were supported by fixed deposit receipts and bank confirmations, and the investigations that followed (by the Central Bureau of Investigation, by the United States Securities and Exchange Commission, which fined the firm 6 million dollars in 2011, and by SEBI, whose order of January 2018 barred Price Waterhouse network firms from auditing listed companies for two years) turned on one point of procedure. Had the auditors obtained those confirmations directly from the banks, or had the documents reached them through the company? An auditing standard is a long text. The question it came down to in Hyderabad was who had carried the envelope.

I remember the following March, the close of the 2008–09 financial year, better than most. Clients who had never asked about audit procedure telephoned to ask whether their own bank confirmations went direct. A tiles distributor in Kothrud, whose turnover was perhaps ₹12 crore, wanted to know whether his auditor would now demand originals of every deposit receipt. I told him the auditor should always have done so. He was quiet for a moment.

'Madam, Satyam had originals also.'

The Answer in Sections

The Companies Act, 2013 received the President's assent on 29 August 2013, and most of its substantive provisions came into force on 1 April 2014. It runs to 470 sections and seven schedules. Read in isolation it can look like a lawyer's attempt to anticipate everything. Read beside the Satyam letter it looks more like a list of replies.

Take the auditors first. Section 139(2) forbids a listed company, and certain larger unlisted ones, from appointing an individual as auditor for more than one term of five consecutive years, or an audit firm for more than two such terms. Rule 5 of the Companies (Audit and Auditors) Rules, 2014 extends the rule to unlisted public companies with paid-up capital of ₹10 crore or more, private companies with paid-up capital of ₹50 crore or more, and any company with public borrowings of ₹50 crore or more. A five-year cooling-off period follows. The reasoning is plain: familiarity is the condition in which envelopes stop being checked. Section 144 bars the auditor from providing a list of other services, including accounting and investment advisory work, to the company it audits. Section 143(12) obliges the auditor who has reason to believe a fraud is being committed by officers or employees to report it to the central government where the amount is ₹1 crore or more. The Act now asks the auditor a question the 1956 Act never put so bluntly: what did you suspect, and whom did you tell?

Then the board. Section 149(4) requires every listed public company to have at least one-third of its directors independent, and Section 149(6) defines independence by a series of exclusions (no pecuniary relationship of a disqualifying kind, no recent employment, no relatives in the promoter group). Schedule IV sets out a code of conduct that reads, in places, like a rebuke addressed to the December 2008 meeting. Independent directors are to 'seek appropriate clarification or amplification of information' and to 'ascertain and ensure' that the company has adequate vigil mechanisms. Section 177 places related-party transactions before an audit committee with an independent majority, and Section 188 requires prior approval, in some cases by shareholders with the interested party barred from voting. Maytas would have had to pass through all of it.

The Act is also careful, in Section 149(12), to limit an independent director's liability to acts committed with their knowledge attributable through board processes, or with their consent or connivance, or where they failed to act diligently. That is a statutory floor. Whether a director reads the bank confirmations personally is a governance ceiling, and I advise my larger clients' independent directors to ask for them. The law does not require it.

Fraud itself acquired a definition. Section 447 defines it broadly, to include acts and omissions intended to gain undue advantage or to injure the interests of the company, its shareholders or creditors, whether or not any wrongful gain or loss actually results. Where the fraud involves at least ₹10 lakh or one per cent of turnover, whichever is lower, the punishment is imprisonment of six months to ten years and a fine of at least the amount involved, up to three times that amount. Where public interest is involved, the minimum term is three years. Section 447 is what the Act points to when it wants to show its teeth.

The Serious Fraud Investigation Office had existed since 2003, set up by executive resolution after the Naresh Chandra Committee's report on corporate audit. It had investigated Satyam without any statute of its own. Section 211 gave it one, and Section 212 gave it powers of arrest. Sahara's answer is quieter and sits in Section 42 with Rule 14 of the Companies (Prospectus and Allotment of Securities) Rules, 2014: a private placement may now be offered to no more than two hundred persons in a financial year. Sahara's two companies had placed their convertible debentures with roughly three crore investors and called it private. The Supreme Court disagreed on 31 August 2012.

Nearly every section I have mentioned was written for a company that had already done something wrong. A founder in Baner, two weeks from incorporation, put it to me more briefly when I showed him the index.

'So which of these is for us?'

Section 7. And the reservation form that comes before it.