The Companies Act 1956 was not repealed on a day. It was dismantled in stages over nearly three years, section by section, notification by notification, while parts of it continued to govern real companies, real proceedings and real disputes long after its replacement had received presidential assent. That is the first thing to understand about the replacement of India’s principal company law statute, because almost every short account of it gets the mechanics wrong. There was no morning on which practitioners woke to a new code and a dead one. There was a long overlap in which two statutes stood side by side, each governing a different slice of corporate life, and in which the answer to the elementary question “which Act applies” depended on the subject matter, the date the proceeding began, and in some corners of the law on provisions of the old statute that were still operating years after the repeal.

The second thing to understand is why replacement, rather than one more amendment, was the chosen route at all. By the time the decision was taken, the Companies Act 1956 had been amended twenty-four times. It had absorbed a committee-driven overhaul in 1988, a governance overhaul in 2000, an electronic filing overhaul in 2006, and in 2002 an amendment so ambitious that it attempted to rebuild the adjudicatory architecture of Indian company law in a single stroke. That 2002 amendment created a National Company Law Tribunal on paper, abolished two existing bodies on paper, and inserted a complete code for the revival of sick industrial companies on paper. Almost none of it was ever brought into force. The tribunal provisions were challenged, held defective in their design of appointments and qualifications, and never notified; the tribunal itself was constituted only in 2016, under the successor statute, fourteen years after Parliament first created it. A statute whose most recent major surgery had failed to take effect was a statute that amendment had demonstrably failed to fix. The nine years between the expert committee that designed the replacement in 2005 and the final transfer of proceedings in 2014 and after are the record of that conclusion being reached, resisted, legislated and then slowly executed.

This article explains why the Companies Act 1956 ended and how its replacement was carried out. The statute that succeeded it is treated in full in the pillar account of the Companies Act 2013 complete guide, and the provision-by-provision comparison between the two Acts belongs to a companion article. The subject here is narrower and, for anyone who wants to understand how Indian legislation actually changes, more instructive: the failure of the old Act, the argument that amendment could no longer repair it, the committee blueprint for its successor, the three bills and the scandal that ran through the passage years, and the repeal and savings machinery by which a repealed statute continued to decide cases while its replacement took hold.

What the Companies Act 1956 was

The Companies Act 1956 was enacted by the Parliament of independent India as Act 1 of 1956 and brought into force on 1 April 1956. It was a consolidating and amending statute: it repealed the Indian Companies Act 1913 and the patchwork of amendments that had accumulated around it, and restated the law of incorporation, management, finance and dissolution of companies in a single text. That text was, by the standards of any legislature, enormous. As enacted it ran to 658 sections arranged in thirteen parts, with fifteen schedules. It dealt with everything from the memorandum of association to the winding up of companies, from the appointment of auditors to the investigation of a company’s affairs by government inspectors, from the management of companies by managing agents to the protection of depositors. It was for decades described, accurately, as one of the longest statutes on the Indian book.

Its intellectual basis was Indian, not colonial, a point worth making because the Act is routinely misdescribed as a British inheritance. The drafting basis was the report of the Company Law Committee chaired by Homi J. Bhabha, appointed in 1950 and reporting in 1952, which examined the working of the 1913 Act and the English Companies Act 1948 and recommended a distinctively Indian settlement: fuller disclosure in prospectuses and accounts, tighter control over managerial personnel and the managing agency system, stronger government powers of investigation and intervention, and protection for minority shareholders against the controlling groups that dominated Indian industry. The 1956 Act carried that settlement into law. Its philosophy was supervisory. The state stood inside the company: approvals were required for managerial appointments and remuneration, for inter-corporate loans and investments, for a range of transactions that in other jurisdictions were left to boards and shareholders. The permission culture this created was not an abuse of the Act but its intended operation, and it generated a distinctive administrative world. The Department of Company Affairs, later the Ministry of Corporate Affairs, sat at the centre; beneath it the Registrar of Companies in each state kept the register and received the filings; the regional directors exercised delegated powers of approval and appeal; and a profession of company secretaries and chartered accountants grew up whose core skill was navigation, knowing which approval was needed, in which office, on which form, with which fee. The Act’s notorious density was, for this world, a kind of job security, and any account of why replacement took nine years should register that a simpler statute threatened real livelihoods built on complexity, even though no committee could say so in terms. Reform of company law in India was never only an argument about texts. It was also a negotiation with the administrative and professional settlement the old text had created, and the replacement’s eventual shape, simpler in the statute and denser in the rules and forms beneath it, bears the marks of that negotiation: the complexity was not abolished, it was relocated downward, from Parliament’s text to the ministry’s rule-book, where the navigating profession could continue to practise its skill on new material. That philosophy fitted an economy of scarcity, licensing and concentrated family ownership. It fitted less well, decade by decade, the economy India became.

The structure of the Act followed its philosophy. The early parts dealt with incorporation, the memorandum and articles, prospectuses and share capital. The long central part dealt with management and administration: directors, meetings, accounts, audit, dividends, and the registers and returns through which the Registrar of Companies supervised compliance. Later parts dealt with compromises and reconstructions, with the prevention of oppression and mismanagement, with government companies, and finally with winding up, which occupied a substantial share of the statute and routed the death of companies through the High Courts sitting as company courts. Around the statute grew an administrative apparatus: the Registrars of Companies in each state, the regional directors, and from the early 1990s the Company Law Board, which took over a growing share of the department’s quasi-judicial work. To administer the Act was to administer a paper empire of filings, approvals and inspections, and its sheer size meant that even the professionals who worked with it daily navigated it by habit and index rather than by comprehension of the whole.

Two features of the Act as enacted matter for everything that follows. The first is that it was prescriptive in form: the obligations of companies were written into the statute itself in fine detail, so that every change in commercial practice, every new financial instrument and every shift in accounting method required Parliament to amend the Act before the law could move. The second is that it concentrated adjudication in the High Courts and in departmental bodies rather than in a specialised forum. Company petitions for winding up, schemes of arrangement, reductions of capital and oppression complaints went to the company judge of the High Court, whose docket they shared with every other species of litigation. Both features were defensible in 1956. Both became, over fifty years, the load-bearing weaknesses of the statute.

The world the 1956 Act was built for, and the one it had to govern

To see why the Act’s structure became a liability, it helps to hold in view the economy for which it was designed and the economy it ended up governing. The committee that designed it worked in the early 1950s, when Indian industry consisted of a few hundred public companies dominated by managing agency houses, when capital was scarce and raised from a narrow investing public, when the state’s development strategy assigned the commanding heights of the economy to public sector enterprises, and when the joint stock company was as much an object of suspicion as an instrument of growth. The Act’s controls were calibrated to that world. Managerial remuneration was capped and approved because managers answerable to controlling families had looted companies in the 1940s. Inter-corporate loans and investments were controlled because agency houses had used the funds of one company to entrench themselves in another. Government companies received their own part of the statute because the state expected to be the largest company promoter in the country, as indeed it became. The managing agency system itself, the institution through which a firm managed a stable of companies for a commission on profits, was regulated in detail rather than abolished, because in 1956 it was still the normal way Indian industry was run; it was abolished only at the end of the 1960s, when its abuses had outlasted its usefulness.

The economy the Act had to govern by the end of its life was unrecognisable against that template. After the liberalisation of 1991, the licensing rationale for the approval culture evaporated, but the approvals remained in the text. India acquired a modern securities market with a statutory regulator, dematerialised trading, foreign institutional investors holding large stakes in listed companies, private equity, employee stock options, cross-border listings and a services economy in which a company could be incorporated in a day and operate in forty jurisdictions. The number of companies on the register grew from tens of thousands to well over a million registrations across the Act’s life. None of this made the 1956 Act’s core concepts obsolete: incorporation, limited liability, the board, the audit and the general meeting are permanent furniture of company law. What it made obsolete was the Act’s method, which was to write the operating instructions of the corporate sector into primary legislation in exhaustive detail and to change them only by parliamentary amendment. A statute that must be amended to accommodate the postal ballot, the dematerialised share and the electronic filing is a statute condemned to permanent arrears, and the arrears are what the next section counts.

The amendment treadmill

The standard response to those weaknesses, for half a century, was amendment. The count most often recorded by the National Foundation for Corporate Governance, the body the Ministry of Corporate Affairs itself promoted to spread governance practice, is twenty-four amendments between 1956 and the final amendment in 2006. Some were minor. Several were structural, and the sequence of the structural ones tells the story of the Act being repeatedly rebuilt inside its own frame.

The Act’s supervisory machinery was tested early, and the test shaped the first decade of amendment. Within months of the Act’s commencement, the government appointed a commission of inquiry under Justice Vivian Bose into the affairs of the Dalmia Jain group, whose interlocking companies had been run through the managing agency structure the Act had chosen to regulate rather than abolish. The commission’s work, which reported in the early 1960s, documented the use of investment companies and inter-corporate transfers to siphon funds on a large scale, and its findings fed directly into the tightening amendments of that decade and into the creation of administrative machinery to watch company management between filings. The episode matters to the replacement story because it fixed the pattern of the Act’s whole life: a scandal would expose a structural weakness, a committee or commission would diagnose it, an amendment would follow, and the amendment would strengthen the department’s hand rather than redesign the system. Scandal, committee, amendment: the rhythm repeated for fifty years, through the Sachar exercise after the excesses of the 1970s and through the governance amendments after the securities scams of the 1990s. The replacement project of the following decade is best understood as the moment the rhythm was deliberately broken, when the diagnosis offered by the committees stopped being “strengthen the machinery” and became “the machinery is the problem”.

The amendments of the 1960s tightened the supervisory model: restrictions on managerial remuneration, controls on inter-corporate investment, the abolition of the managing agency system at the end of that decade. The next great wave followed the report of the High-Powered Expert Committee on Companies and the Monopolies and Restrictive Trade Practices Act, chaired by Justice Rajinder Sachar, which reported in 1978. The Companies (Amendment) Act 1988, enacted a decade after the report, was the largest single surgery performed on the statute in its middle life: it recast provisions on deposits, on the powers of the Company Law Board, on inspection and investigation, and on the protection of investors, and it is the point at which the 1956 Act recognisably became a different statute from the one the Bhabha Committee had designed. It also introduced the compounding of offences, the mechanism by which a company or its officers could settle an offence on payment without a prosecution, which was an admission written into the statute itself that the criminal process had become the wrong instrument for most of the defaults the Act created. Compounding grew over the following decades into the routine exit from the Act’s penal provisions, used in thousands of cases, and its history is a parable of the whole system: faced with a sanction structure that did not fit the conduct it governed, the system did not redesign the structure but built a settlement valve beside it, and the valve became the main channel. The Irani Committee’s later proposal, that procedural defaults should attract monetary penalties imposed administratively rather than offences settled by compounding, was an attempt to design honestly what compounding had improvised. That the proposal was only partly enacted in 2013, and that compounding crossed into the new Act and continued its career there, shows how deeply the improvisation had settled into the expectations of everyone who used the system. Yet even its authors treated it as renovation of a standing structure. The registers, the approval culture and the court-centred adjudication all survived it.

The amendments of the reform era changed the Act’s direction without changing its architecture. The Companies (Amendment) Act 2000 wrote the vocabulary of corporate governance into the statute: it required audit committees for public companies above a prescribed capital, formalised the responsibilities of directors in the board’s report, introduced the postal ballot so that specified resolutions could be carried by shareholders voting by post, and adjusted the regime for buy-back of shares and for debentures. The Companies (Amendment) Act 2006 then changed the medium of the Act’s administration rather than its substance: it provided for director identification numbers and for the filing of documents in electronic form, and it underpinned the MCA21 programme through which the registry moved from paper files in state offices to a single electronic system. In the space of six years the Act had been asked to become a governance code and then a digital code. Each time, the new material was bolted onto a frame designed for neither.

Twice in the 1990s the government of the day attempted the more radical course and failed. Comprehensive amendment bills introduced in 1993 and again in 1997, each intended as a thorough recasting of company law, never completed their passage, and the effort lapsed back into incremental amendment. The pattern matters. By the time the replacement project began in earnest, the proposition that the 1956 Act could be modernised by amendment was not an untested hypothesis. It had been tested in 1993, in 1997, in the governance amendments of 2000, and most decisively in 2002, and each test had either failed in Parliament or failed in execution. The case for replacement was built less on the Act’s age than on this record.

The vanishing companies and the investor the Act could not protect

Between the governance amendments and the replacement, the old Act suffered a failure of a cruder kind, and it belongs in the catalogue because it shaped the investor protection provisions of the successor. During the primary market boom of the mid 1990s, when the office of the Controller of Capital Issues had been abolished and public issues were freed, a large number of companies raised money from the public and then, in the phrase that entered the official vocabulary, vanished: the promoters became untraceable, the registered offices were found empty, the shares became worthless paper and the investors, mostly small and in small towns, had no remedy that worked at the speed of their loss. Successive governments appointed committees to identify the vanished companies and to pursue their promoters, prosecutions were launched under the 1956 Act’s penal provisions, and the episode ran for years through the late 1990s and the decade following.

The legal significance of the episode is diagnostic. Nothing in it turned on an absence of offences in the statute; the Act criminalised false prospectuses, fraud and default in plenty. What failed was the machinery of pursuit: a department organised for filings and approvals, registrars in the states with tiny enforcement staffs, and a prosecution process that ground forward for years while the money and the men dispersed. The reform conclusions drawn from the episode fed the successor statute at several points: the statutory fraud office, the disgorgement and class action provisions, the tightening of the deposits regime so that the public’s money could not be taken by companies outside a disciplined framework, and the investor education and protection fund. More broadly, the episode taught the reform generation the difference between a statute that prohibits and a state that enforces, which is the same difference the tribunal story teaches in the adjudicatory field. The 1956 Act, by the end, prohibited nearly everything and enforced selectively and slowly, and each scandal of its last decades, vanishing companies, plantation and finance company collapses absorbed into its enforcement record, Satyam at the summit, restated the same deficiency in a larger font. An amendment can add an offence. The deficiency was never a shortage of offences.

Governance by listing agreement: the Act’s authority leaks away

One measure of the 1956 Act’s exhaustion, rarely given its full weight in accounts of the replacement, is that during the last fifteen years of its life the centre of gravity in Indian corporate governance moved out of the Act altogether. The Securities and Exchange Board of India, armed with the listing agreement between exchanges and listed companies, built the operative governance code for listed companies contractually, because the listing route could be revised by a regulator’s circular in months while the statutory route waited on Parliament for years. The committee chaired by Kumar Mangalam Birla, appointed by the securities regulator, reported in 1999, and its recommendations entered the listing agreement as the celebrated Clause 49 in 2000: independent directors, audit committees of the board, certification of accounts, disclosure of related party dealings and a compliance report, all imposed on listed companies not by the Companies Act but by the terms on which their shares were traded.

Why did corporate governance enter through listing rules instead of the Act?

Because the listing agreement could be rewritten by the regulator as practice developed, while every change to the statute required a parliamentary amendment; for the fifteen years before the replacement, the governance obligations that actually bit on listed companies lived in Clause 49 of the listing agreement rather than in the Companies Act that nominally governed them.

The consequences for the statute were double. First, the Act and the listing code diverged: audit committees existed in both after 2000, with different compositions and different consequences, and a listed company’s governance life was lived in the space between two instruments administered by two authorities, the department and the regulator, whose jurisdictions overlapped without a clean boundary. Second, and more damaging to the Act’s authority, the divergence taught the market a lesson about where Indian company law was actually made. A code that must watch its central subject being regulated elsewhere is a code in liquidation, whatever its formal status. The reform committees of the period all registered the point: the Naresh Chandra Committee’s work on audit and governance was addressed as much to the statutory framework as to the listing one, and the Irani Committee recommended that the new Act state a single governance framework for companies with the regulator’s code aligned to it. The Companies Act 2013 attempted exactly that absorption, writing independent directors, audit committees and board evaluation into the statute and leaving the listing obligations to operate as an overlay for listed companies. That the dual structure persists in a new form is a criticism of the successor’s execution; that the statute had already forfeited the field by 2005 is a fact about the predecessor that no amendment of the predecessor was ever likely to reverse, because the forfeiture was caused by the amendment process itself, by its slowness relative to a regulator’s circular.

The costs of the palimpsest

The amended Act also failed its users in a quieter way, through the character of its obligations and sanctions. A very large share of the 1956 Act’s requirements were procedural: filings, registers, notices, approvals, each with a timeline, and a very large share of its offences were defaults in that procedure, punishable with fines and in a number of provisions with imprisonment of officers. The result, documented by every committee that examined the Act’s working, was a compliance regime in which the criminal process was routinely engaged by technical default and routinely exited by compounding, the statutory mechanism by which offences could be settled on payment. The Sachar Committee had worried at the problem in 1978; the working group that preceded the abortive 1997 bill had proposed wholesale decriminalisation of procedural defaults; the Irani Committee returned to it, recommending that the new law grade its responses, reserving criminal sanction for fraud and dishonesty and handling procedural default through monetary penalties adjudicated administratively. The complaint was not that the Act was strict. It was that its strictness was misallocated, severe on the missed filing and comparatively helpless before the determined fraudster, whose conduct the Act reached only through investigation provisions that were slow, departmental and rarely used to conviction.

This is the context in which the fraud provisions of the successor statute should be read, and it explains a feature of the replacement that pure comparison tables obscure. The 2013 Act did not merely reorganise the old Act’s contents. It attempted to re-found the sanctioning philosophy: a defined offence of fraud carrying serious punishment, a fraud investigation office with statutory powers of arrest and prosecution, and, on paper, a penalty regime for the routine defaults. That the attempt was only partly carried through, and that the decriminalisation the committees recommended arrived in instalments through amendments of the new Act rather than in the Act itself, belongs to the successor’s record. For the predecessor, the relevant fact is simpler: by the time the replacement was designed, the 1956 Act’s sanction structure was itself one of the exhibits against it, and sanction structure is not the kind of defect a schedule of amendments repairs, because it is distributed across hundreds of provisions rather than located in any one of them.

Why amendment stopped being enough

The decision to replace rather than amend rested on a diagnosis with several distinct parts, and it is worth separating them, because the replacement statute answered each part differently and in one case barely answered it at all. The diagnosis was that the Act’s defects were not defects of detail that a schedule of amendments could cure. They were defects of structure: the way the statute stored its rules, the forums it sent disputes to, the machinery it lacked, and the remedies it gave to the shareholders it was supposed to protect.

Where did the idea of a single company law tribunal come from?

It came from the committee chaired by Justice V. Balakrishna Eradi, appointed to examine the law relating to insolvency and the winding up of companies, which reported in 2000 and recommended that company jurisdiction scattered among the High Courts, the Company Law Board and the sickness machinery be gathered into one specialised tribunal with an appellate tribunal above it.

The Eradi Committee’s report was the product of a specific scandal of delay. Winding up under the 1956 Act was conducted by the High Courts, and a company petition could spend years moving between the stage of admission, the appointment of the official liquidator, the realisation of assets and the settlement of claims, while the assets themselves decayed. The committee’s answer was institutional rather than textual: a single tribunal, staffed by judicial and technical members, exercising the company jurisdiction of the High Courts together with the powers of the Company Law Board and the jurisdiction over sick companies, so that the life and death of a company would be supervised by one forum built for the purpose. Almost everything that later happened to the tribunal idea, its enactment in 2002, its destruction in litigation and its eventual constitution in 2016, follows from that recommendation. It also supplies the cleanest illustration of the difference between amending a statute and replacing one, because the recommendation required not new words on old pages but a new institution, and institutions cannot be created by drafting alone.

Why did winding up a company under the 1956 Act take so long?

Winding up took so long because the Act routed it through the ordinary courts and an overextended official machinery: the company judge of each High Court heard petitions alongside a general docket, the official liquidator handled realisation and claims with thin staffing, and every interlocutory dispute could travel upward on appeal, so a liquidation measured in years was routine.

The consequences compounded. While a winding up crawled, secured creditors enforced their security outside the process where they could, workers waited for dues that ranked in the statute but arrived in practice only if assets survived the delay, and the premises and plant of the dead company deteriorated into scrap. The Eradi Committee treated this not as a failure of individual courts but as a design failure: a statute that made the generalist court the liquidator of industry had built delay into the definition of the process. It is telling that when the replacement of the winding-up regime finally came, it came not from the Companies Act 2013 itself, whose own provisions on the subject waited years for notification, but from a separate statute, the Insolvency and Bankruptcy Code 2016, which took corporate insolvency and liquidation out of the company law frame altogether and gave them a timed process of their own. The winding-up failure of the 1956 Act was, in the end, answered by subtraction.

The other failures were of a different kind. The Act contained no independent regulator for the audit profession. Auditors were appointed by shareholders, disciplined through the professional institute established under the Chartered Accountants Act 1949, and overseen by the department only at the level of individual investigations. When accounting failure became the central corporate scandal of the age, there was no public authority charged with setting and enforcing audit standards at arm’s length from the profession itself. The Naresh Chandra Committee, which reported on corporate audit and governance in December 2002, recommended an independent oversight mechanism for auditors; the 1956 Act had no shelf on which to put it. The remedy for oppression and mismanagement, contained in the sections of the Act that allowed members to petition the Company Law Board, was real but narrow: there was no class action by which investors as a group could pursue a wrong done to them collectively, no statutory route for a depositor or debenture holder class, and no derivative mechanism worth the name for wrongs done to the company itself. Minority shareholders in closely held companies used the oppression remedy heavily and often effectively, but the dispersed investor in a listed company, the figure the reform era cared most about, had disclosure rules and little else.

The audit gap deserves its mechanism to be stated, because it is the failure that the scandal of 2009 converted from a committee observation into a public fact. Under the 1956 dispensation, the auditor of a company was appointed at the annual general meeting, reported to the shareholders, and was removable by them, but in a market of controlling shareholders the appointment was in practice the management’s gift, and the auditor’s economic future depended on the goodwill of the people whose accounts he was examining. Oversight of the profession lay with the Institute of Chartered Accountants of India, a statutory body that both represented chartered accountants and disciplined them, so that the investigator of a failed audit and the professional home of the auditor under investigation were the same institution. The department could order investigation of a company’s affairs, and the board could be superseded in extreme cases, but there was no authority whose standing job was to inspect audit quality, to require firms to demonstrate their systems, and to sanction failure before it matured into a collapsed company. The Naresh Chandra Committee recommended an independent board for audit oversight in 2002. The recommendation was not implemented under the old Act, and the reason usually given, a decade of negotiation with the profession, is itself a datum about the old regime: a statute whose enforcement depended on departmental stamina had no stamina left for a fight with an organised profession. When Satyam’s accounts proved to have been audited for years without the auditors detecting that the bank balances did not exist, the absence of the oversight body stopped being an institutional nicety. The 2013 Act’s creation of the National Financial Reporting Authority is the replacement’s single clearest answer to a named failure, and the nine years between the recommendation and the enactment, and the further five before the Authority was constituted in 2018, measure the distance between amending a text and building the institution the text describes.

The minority remedy gap has an equally concrete illustration, and it also runs through Satyam. The investors who lost money in Satyam included holders of its American depositary shares, who sued in the United States under its securities class action procedure and obtained a settlement of 125 million dollars from the company. Investors who had bought the same company’s shares on Indian exchanges had no corresponding vehicle: the 1956 Act gave them an oppression petition designed for shareholders in closely held companies, a criminal process they did not control, and the regulator’s enforcement discretion. A legal system in which the foreign holder of an Indian company’s paper has a collective remedy and the domestic holder does not has its priorities inverted, and the inversion was structural, not accidental: class actions had been recommended in the reform literature and resisted, on the stated fear of speculative litigation, through successive amendment cycles. The 2013 Act’s class action provision, whatever its subsequent career, exists because the replacement method allowed a remedy to be introduced as part of a general settlement rather than fought clause by clause through an amending bill in which every interested party could isolate and kill it. That is a genuine, if underappreciated, advantage of replacement as a legislative technique: bundling changes the politics of each individual change.

Above all, the Act had become administratively indigestible. Six hundred and fifty-eight sections, many of them amended repeatedly, cross-referred to rules, forms and departmental circulars accumulated over five decades; the compliance burden fell on every company regardless of size, because the statute drew few distinctions between the listed multinational and the family firm except by exemption notification. Every reform proposal of the period begins with the same observation, that the Act’s size had become a defect in itself, defeating the clarity that a code is supposed to provide and multiplying the points at which a company could default without any intent to do wrong.

The committee blueprint for a new Act

The replacement was designed before it was drafted, and designed by a committee. In December 2004 the Ministry of Company Affairs set up an expert committee under the chairmanship of Dr J. J. Irani, a director of Tata Sons, to advise on a new company law. The Irani Committee reported on 31 May 2005, and its report is the intellectual origin of the Companies Act 2013 in a way that can be traced clause by clause. The committee’s terms of reference included bringing about compactness by reducing the size of the Act and removing redundant provisions, enabling easy and unambiguous interpretation by recasting the law, and providing greater flexibility in rule-making so that the law could respond to changing business models without returning to Parliament for every adjustment. Those three instructions, compactness, clarity and flexibility, became the architecture of the successor statute.

The substance followed the architecture. The Irani Committee recommended a principles-based statute: the Act itself would state the governing principles and the hard obligations, while the detail that changes with practice, the forms, thresholds, procedures and timelines, would move into rules made by the executive and laid before Parliament. It recommended a single governance framework across company types with calibrated obligations, easier incorporation, recognition of the one person company, stronger disclosure and board accountability, an audit regime with independent oversight in the line of the Naresh Chandra recommendations, a serious fraud investigation office with statutory standing, better protection for minority interests, and the tribunal system the Eradi Committee had proposed. It also recommended, crucially, that the new law be enacted as a new statute rather than as an amendment of the old, because the recasting it proposed touched the arrangement of the whole.

The argument between amendment and replacement was therefore not a slogan but a drafting question. If most of the operative detail was to move out of the Act into rules, if dead provisions were to be dropped wholesale, if new institutions with no counterpart in the old text were to be created, and if the surviving provisions were to be renumbered and regrouped by subject rather than by the accidents of fifty years of amendment, then the honest legislative vehicle was a fresh Act with a repeal and savings clause, not a two-hundred-clause amendment bill that would leave practitioners collating two texts. The committees had also supplied the political cover for that choice: the Eradi report for the tribunal and winding up, the Naresh Chandra report for audit and governance, and the Irani report for the whole design. What remained was to get a bill through Parliament, which proved to be the slowest part of the enterprise.

What the Irani design asked of the state

The Irani report deserves a closer reading than it usually receives in this story, because the replacement that eventually emerged can be judged only against the design that justified it, and because the design placed demands on the Indian state quite different from the demands the 1956 Act had placed. The old Act demanded stamina: a department able to process approvals, filings and inspections at industrial scale. The Irani design demanded restraint and drafting discipline: a state willing to state principles in the statute, trust boards and shareholders with the conduct those principles governed, police the boundary through disclosure and a small number of well-aimed sanctions, and keep the rule-book beneath the Act current without drama. It is a more sophisticated demand, and the history of the replacement is in part the history of the state accepting the design rhetorically while struggling with its discipline in practice.

Three elements of the design show the intention most clearly. The first is the reallocation of detail. Under the 1956 Act, the fee for a filing, the form of a register, the timeline for a notice and the contents of a report were statutory or near-statutory, frozen at the moment of enactment. Irani proposed that the Act fix the obligation and its outer limits while rules prescribed the mechanics, so that a change in the fee or the form required a notification, not a session of Parliament. The second is differentiation by scale. The old Act’s exemptions for private companies had grown by accretion, concession by concession, until the boundary between the regulated and the lightly regulated was a matter of circulars. Irani’s framework proposed instead that obligations be calibrated openly to the size and public character of the company, with the smallest companies, including the one person company the report recommended recognising, carrying a genuinely lighter load. The third is exit. The old Act made it easy to incorporate a company and extraordinarily hard to bury one, with the result that the register accumulated defunct companies by the hundred thousand, struck off slowly if at all. The design treated easy exit, through strike-off and a functioning liquidation process, as part of the law of entry, on the reasoning that a register full of the dead misinforms everyone who relies on it.

Against that design, the committees’ consensus had one notable gap, and it foreshadows the largest criticism of the Act that emerged. None of the committees resolved the question of volume: how much obligation, in total, a company should carry. Irani’s principles-based method was expected to deliver the reduction by itself, since principles compress where prescriptions sprawl. Whether that expectation was realistic, whether a principles-based statute administered by a cautious department and audited by a risk-averse profession stays principles-based in practice, is the question the successor statute’s record poses. It is taken up in the assessment below; it is flagged here because it is the point at which the replacement’s design and the replacement’s execution part company, and an honest history has to keep the two apart.

Three bills, nine years

The Companies Bill 2008, introduced in the Lok Sabha on 23 October 2008, was the first attempt to enact the Irani design. It never reached a vote. When the fourteenth Lok Sabha was dissolved ahead of the general election of 2009, the bill lapsed with it, as bills pending in the House do. Between the lapse and the introduction of the next bill, the politics of company law changed entirely, because on 7 January 2009 the chairman of Satyam Computer Services, B. Ramalinga Raju, wrote to his board confessing that the company’s accounts had been falsified on a scale without Indian precedent: of the cash and bank balances shown in the books, Rs 5,040 crore did not exist, and the total falsification, counting accrued interest that had never been earned, an unrecorded liability and inflated receivables, came to Rs 7,136 crore. The government superseded the Satyam board within days and the company was sold to a rescuing investor within months, but the legislative aftershock lasted years.

The convenient narrative is that Satyam produced the new company law, and the narrative is wrong in a way this series insists on correcting wherever it appears. The reform process predated the scandal by years: the Irani Committee had reported in 2005, the bill embodying its design had been introduced in 2008, and the architecture of the 2013 Act, the compact statute with detail in rules, the tribunal, the fraud office, the audit oversight, was settled before anyone outside Satyam’s finance function knew the accounts were false. What the scandal changed was narrower and real. It changed the political weight carried by the audit and fraud provisions, it made the case for an independent audit regulator and a statutory fraud investigation office nearly impossible to argue against in committee, and it raised the priority of the bill in a crowded legislative programme. A scandal can accelerate a statute and sharpen particular clauses. The record does not support the claim that it authored one.

There is an irony in the Satyam investigation that belongs in any honest history of the old Act. The fraud was investigated, in its corporate law dimension, with the 1956 Act’s own tools: the Serious Fraud Investigation Office, which conducted the inquiry, had been created in 2003 as an executive office and drew its coercive powers from the investigation provisions of the very statute whose replacement the scandal was about to accelerate. Those provisions, empowering the Central Government to appoint inspectors, to compel the production of documents and to prosecute on the inspectors’ report, were among the strongest in the Act, and in the Satyam matter they worked: the investigation was completed, prosecutions were launched, and convictions followed in the trial court in 2015. The episode therefore cuts against any simple story of a useless statute. The 1956 Act’s investigative spine was serviceable. What it lacked was everything around the spine: the preventive oversight that might have detected the fraud before the confession, the collective remedy for the investors it destroyed, and the specialised forum to try the corporate consequences quickly. Replacement was aimed at the surroundings, not at the spine, and the new Act duly carried the investigation machinery across in strengthened statutory form, giving the fraud office the standing the old Act had never given it.

The Companies Bill 2009 was introduced in the Lok Sabha on 3 August 2009 and referred to the Standing Committee on Finance, which reported on 31 August 2010 with recommendations extensive enough that the government chose to withdraw and recast rather than amend. The recast text, the Companies Bill 2011, was introduced in December 2011 and sent back to the same committee.

What did the Standing Committee on Finance change in the Companies Bill?

The committee’s two examinations, of the 2009 bill and the 2011 bill, reporting in August 2010 and June 2012, pressed the text toward stronger investor protection and director accountability, questioning the dilution of safeguards for depositors and small shareholders and seeking tighter provisions on related party transactions and managerial remuneration, in numbers sufficient that the government twice preferred recasting.

The Lok Sabha passed the bill on 18 December 2012. The Rajya Sabha passed it on 8 August 2013, and it received the assent of the President on 29 August 2013, becoming the Companies Act 2013, Act 18 of 2013: 470 sections in twenty-nine chapters, with seven schedules, and a body of rules beneath it carrying much of the operative detail the Irani Committee had wanted out of the statute. A small textual detail marks the character of what had been enacted. The long title of the Companies Act 2013 describes it as an Act to consolidate and amend the law relating to companies, the same formula of words that had introduced the Act of 1956, which had itself consolidated and amended the law of 1913. Indian company legislation does not, in its own formal language, admit to revolution: each generation’s statute presents itself as the orderly restatement of the last, and the repeal of the predecessor is handled in a miscellaneous section near the end rather than proclaimed in the title. The continuity of formula is not hypocrisy. It reflects the truth, developed later in this article, that a replacement of this kind preserves far more than it destroys, and that its drafters knew it. But it does mean that the scale of the 2013 change has to be read from its machinery and its institutions rather than from its self-description, which is one reason the change was persistently underestimated in its first years by readers who took the long title at its word.

Measured from the constitution of the Irani Committee in December 2004, the replacement had taken eight years and nine months, three bills and two standing committee examinations to enact. The replacement then took a further period, longer than anyone had planned, to execute, and the execution is where the repeal of the 1956 Act becomes genuinely interesting as a piece of legislative craft.

The comparative mirror

India’s replacement exercise was not unique, and the comparison most often drawn in the Indian debate was with the United Kingdom, for the natural reason that the two statutes were cousins: the Indian Act of 1913 and the Act of 1956 both descended from the English companies legislation, and the Bhabha Committee had worked with the English Companies Act 1948 open on the table. The United Kingdom conducted its own company law review from the late 1990s and enacted the Companies Act 2006, a statute larger even than India’s, which restated and reformed the law and was then commenced in stages across two years, with the new provisions phased in from 2007 to 2009 while the Companies Act 1985, the predecessor, was correspondingly disapplied. The parallel is instructive in both directions. It acquits the Indian drafters of the charge that staged commencement was a local eccentricity: the jurisdiction from which the whole tradition derived chose the same technology for the same reason, that a company code cannot be swapped overnight. And it sharpens the criticism available against the Indian execution, because the British staging was accompanied from the outset by a published commencement architecture mapping old provisions to new, while the Indian transition left the corresponding-provisions mapping of September 2013 to a circular and to professional improvisation. The method was sound and internationally orthodox. The documentation of the method was where India fell short, and the shortfall was administrative, not legislative.

The comparison also disciplines the rhetoric of size. The British Act of 2006 is longer than the statute it replaced, and no one in that debate treated length as the measure of failure; the measure was whether the text was organised intelligibly, whether its obligations were proportioned to the companies bearing them, and whether its institutions worked. By those measures the case against the Indian Act of 1956 was never really its 658 sections. It was that its organisation recorded the order of its accidents rather than the logic of its subject, that its obligations were unproportioned as between the largest and the smallest companies, and that its central new institution had been stillborn. Size was the symptom quoted in speeches. The replacement debate, at its best, in the committees, knew the difference, and the successor statute should be judged by the same three measures rather than by its section count, a point the assessment below applies.

Why nine years: the politics of a bill nobody opposed

A puzzle in the chronology deserves attention before the tribunal story resumes it, because it generalises beyond company law. The replacement of the 1956 Act was not, at any point, a seriously contested partisan project. No party campaigned against a new companies Act; industry wanted modernisation; the profession wanted clarity; the committees were consensual and technocratic. Yet the project took from December 2004 to August 2013 to enact, consumed three bills, and survived two dissolutions of the Lok Sabha only by being reintroduced. The explanation lies in the structure of legislative time. Company law reform has no electoral constituency and no deadline. It competes for parliamentary hours against finance bills that must pass, against politically charged legislation that governments need and oppositions want to fight, and against the simple congestion of the Indian parliamentary calendar. A bill that everyone supports and nobody needs this session is the bill most easily postponed, and the Companies Bill was postponed through the life of two Parliaments.

Two interruptions were substantive rather than calendrical. The dissolution of the fourteenth Lok Sabha in 2009 killed the 2008 bill mechanically, by the rule that bills pending in the House lapse with it; the incoming government had to decide afresh whether to adopt its predecessor’s text, and its decision to proceed through a fresh bill and a fresh committee examination, rather than fast-tracking the lapsed text, cost the project roughly three years and bought it the Standing Committee’s extensive revisions. The second interruption was the change of drafting philosophy between bills: the 2009 bill as introduced was found, in committee, to have diluted safeguards the committee wanted restored, and the government’s acceptance of recasting over patching, while creditable, reset the clock again in 2011. The lesson the sequence teaches, and the reason this article spends time on it, is that the failure mode of Indian legislative replacement is rarely defeat. It is drift: a reform everyone endorses, proceeding at the speed of a crowded agenda, while the statute it is meant to replace continues to age underneath it. The nine years were not spent arguing about whether the 1956 Act should be replaced. They were spent not quite getting around to replacing it.

The tribunal that amendment built and nobody could staff

The single most instructive episode in the decline of the 1956 Act is the fate of its own last great amendment. The Companies (Second Amendment) Act 2002, following the Eradi Committee, rewrote the institutional map of company law. It inserted into the 1956 Act new parts providing for a National Company Law Tribunal and a National Company Law Appellate Tribunal, to which the company jurisdiction of the High Courts, the powers of the Company Law Board and the jurisdiction over sick companies were to move. It abolished, prospectively, the Company Law Board and the Board for Industrial and Financial Reconstruction with its appellate authority. And it inserted a self-contained code, running through sections numbered 424A to 424L, for the revival and rehabilitation of sick industrial companies, intended to take that subject away from the sickness statute of 1985.

Almost none of this was ever switched on. The tribunal provisions were made dependent on notification, and before notification came constitutional challenge. The Madras Bar Association attacked the design in the Madras High Court, which in March 2004 held key provisions defective; on appeal, a Constitution Bench of the Supreme Court decided the matter in Union of India v. R. Gandhi, President, Madras Bar Association, in May 2010. The Court upheld Parliament’s competence to create the tribunals and to transfer company jurisdiction to them, which was the point of principle. But it held the provisions governing the qualifications and appointment of members to be defective in ways that went to the tribunals’ independence: persons of insufficient seniority and standing were made eligible to do work that had been the work of High Court judges, and the selection arrangements gave the executive too large a hand. The provisions required correction before the tribunals could be constituted in conformity with the Constitution. The corrections were never made to the 1956 Act. The tribunal chapter of the 2013 Act was drafted to absorb the judgment, was challenged in its turn by the Madras Bar Association, and in 2015 the Supreme Court held that the new Act had repeated part of the old error: the provisions fixing the qualifications of technical members of the Tribunal and the Appellate Tribunal, in Sections 409 and 411, were struck down insofar as they again admitted persons below the seniority the Constitution Bench had required in 2010, and the government was directed to apply the 2010 corrections in making appointments. Only after that second judgment did the executive move, constituting the tribunals the following year with appointments made under the corrected dispensation. The sequence repays attention, because it shows the courts performing, across a decade, the quality control on institutional design that the legislative process had failed to perform for itself: twice Parliament enacted a tribunal, and twice the enactment’s personnel provisions had to be judicially corrected before the institution could safely be born. It also acquits the replacement Parliament of the simplest version of the charge against it. The delay in constituting the Tribunal was not idleness. A tribunal constituted on the 2002 text would have been born unconstitutional, and everyone in the system knew it after 2010. The detailed account of that litigation and of the tribunal’s eventual working belongs to the article on the National Company Law Tribunal and its appellate tribunal; the point for the replacement story is the outcome and the dates. A tribunal created by Parliament in 2002 was constituted on 1 June 2016. Fourteen years elapsed between the legislative decision and the institution, and they elapsed not because the idea was abandoned but because an amendment had built an institution whose design the Constitution, as the courts read it, would not accept, and no government chose to spend the parliamentary time repairing the amendment when a replacement statute was already on the stocks.

The litigation also froze the successor statute in its cradle, a downstream consequence that connects the tribunal story directly to the transition story told in the next section. When the main tranche of the Companies Act 2013 was notified in the spring of 2014, the sections left unnotified were not random. They were, in the main, the sections that depended on the Tribunal or on other unbuilt machinery: the tribunal chapter itself, the oppression and mismanagement provisions that were to move from the Company Law Board to the Tribunal, compromises and arrangements, winding up, the sick companies chapter, special courts and the audit authority. The ministry’s own status notes of the period attribute the withholding of the tribunal-dependent sections to the case pending in the Supreme Court on the Tribunal’s design. So the 2002 amendment’s defect propagated forward a full decade: because the courts were still correcting the design of a tribunal first legislated in 2002, the replacement Act of 2013 spent its first three years with its adjudicatory heart uninstalled, and the repealed Act’s Company Law Board, preserved by the savings proviso, heard the cases instead. Institutional failure, in this story, is not an event but a inheritance, passed from the dying statute to its successor.

The sickness provisions tell the same story in a second register. The Sick Industrial Companies (Special Provisions) Act 1985, Act 1 of 1986, had created the Board for Industrial and Financial Reconstruction to detect sickness in industrial companies and to nurse or bury them; by common consent the board’s proceedings had become a shelter in which defaulting managements could park their companies beyond the reach of creditors, because a reference to the board suspended recovery proceedings. The 2002 amendment’s sick companies code was meant to end that regime inside the Companies Act. It was never notified. Parliament then enacted the repeal of the sickness statute itself, the Sick Industrial Companies (Special Provisions) Repeal Act 2003, and that repeal in turn was left uncommenced for thirteen years, so that the BIFR continued to function under a statute that Parliament had already voted to repeal, precisely because nothing had been put in its place. The repeal was finally brought into force on 1 December 2016, the day the Insolvency and Bankruptcy Code’s corporate insolvency provisions began, whereupon the board and its appellate authority stood dissolved and pending references abated, with companies given a window to move to the tribunal under the Code. The fuller account of that long deferral belongs to the article on the repeal of SICA and the end of the BIFR. For this article the episode does double duty: it is both a failing of the old dispensation, since the sickness jurisdiction was part of the company law landscape the 1956 Act inhabited, and a demonstration that in Indian practice repeal itself is a staged and conditional act, a lesson the drafters of the 2013 Act’s own repeal clause had absorbed.

The repeal itself: Section 465 and the staged commencement

The Companies Act 2013 did not commence on its assent. Its first section came into force at once, on 30 August 2013, and the remaining provisions were left to be brought into force by notification, on different dates for different provisions. That single design choice governed everything that followed, because it converted the replacement of a statute from an event into a process, and it made the repeal of the old Act follow the commencement of the new rather than the calendar.

The repeal clause is Section 465. It provides that the Companies Act 1956 and the Registration of Companies (Sikkim) Act 1961 stand repealed, and then, in the same breath, it qualifies the repeal in ways that kept the old Act alive in defined compartments. Two provisos matter most. Until the date notified for the transfer of matters to the Tribunal under Section 434, the provisions of the 1956 Act governing the jurisdiction, powers and functions of the Board of Company Law Administration, that is the Company Law Board, and of the courts were to continue to apply as if the 1956 Act had not been repealed. And the provisions of Part IXA of the 1956 Act, the producer companies chapter inserted in 2002, were to continue to apply to producer companies as if the Act had not been repealed, until a special law for producer companies was enacted. Section 465 then sets out a savings code of the standard but carefully enumerated kind: everything done under the repealed enactments, every rule, order, appointment, proceeding, penalty and registration, is deemed to have been done under the corresponding provisions of the new Act so far as it is not inconsistent with them; companies incorporated under the old Act continue to be validly incorporated and are governed by the new; registers and funds continue; prosecutions pending in court continue to be heard by that court; inspections and investigations ordered under the 1956 Act continue as if ordered under the new; and matters pending before the Registrar, a regional director or the Central Government are to be concluded under the old Act notwithstanding its repeal. The section closes by preserving the general application of Section 6 of the General Clauses Act 1897, the standing Indian provision on the effect of repeal, so that accrued rights and liabilities survive by both specific and general savings.

Did the Companies Act 2013 repeal the 1956 Act in one step?

No. The repeal operated in step with the commencement of the new Act: as each tranche of the 2013 Act was notified, corresponding provisions of the 1956 Act ceased to operate, and compartments of the old Act, including the Company Law Board’s jurisdiction and the producer companies chapter, were expressly kept alive by Section 465 until their successors were ready.

The tranches came as follows. On 12 September 2013, ninety-eight sections of the new Act were notified, principally definitional and machinery provisions that needed no rules and no new institution. The Ministry of Corporate Affairs then had to clarify, by a general circular issued within the week, that the provisions of the 1956 Act corresponding to the ninety-eight would cease to have effect, and it left the identification of the corresponding provisions to companies and their advisers, an early taste of the interpretive burden the staged method imposed. On 1 April 2014 a further 183 sections came into force, with the principal rules under them, bringing the great bulk of the operating code, incorporation, prospectus and allotment, share capital, management and administration, accounts and audit, into the new regime: from that date the ordinary life of an Indian company was lived under the 2013 Act. What remained under the old dispensation after April 2014 was, significantly, the adjudicatory and terminal machinery: the tribunal provisions, the oppression and mismanagement jurisdiction still exercised by the Company Law Board under the saved provisions of the old Act, compromises and reconstructions still before the High Courts, and winding up, which continued under the 1956 Act in the company courts for more than two further years.

Two further compartments of survival are worth recording because they show the draftsman’s care at the edges. The Limited Liability Partnership Act 2008 had borrowed, by notification, a set of provisions of the Companies Act 1956 and applied them to limited liability partnerships; Section 465 therefore provided that those borrowed provisions would continue to apply to partnerships as if the 1956 Act had not been repealed, until a corresponding notification applied the new Act’s provisions instead, so that the repeal of the companies statute would not silently strip a different statute of its working parts. And the Registration of Companies (Sikkim) Act 1961, the separate incorporation law of Sikkim which the 1956 Act had never displaced because of the state’s distinct constitutional position, was repealed by the same section, closing a survival from the era of the princely accessions and bringing the whole territory of India under one incorporation code for the first time. Repeals are usually read for what they kill. This one is better read for what it was careful not to kill by accident.

The compartments were then closed one by one as their successors became real. The Tribunal was constituted on 1 June 2016 and the Company Law Board dissolved the same day, with pending board matters transferred to the Tribunal; the corresponding sections of the new Act, including those on oppression and mismanagement, were notified with it. Compromises, arrangements and capital reductions followed the tribunal route in the notifications of that year. Winding up moved last and by a side door: the Insolvency and Bankruptcy Code 2016, which received assent in May 2016, deleted the 2013 Act’s own dormant chapter on sick companies, supplied the insolvency and liquidation process the system had been waiting for, and from December 2016 the Insolvency and Bankruptcy Code 2016 began receiving the winding-up business of the old Act, with petitions at an advanced stage in the High Courts staying where they were and the rest transferring to the Tribunal under the transfer rules. Producer companies, the smallest and strangest survival, stayed under Part IXA of the repealed Act until the Companies (Amendment) Act 2020 wrote a producer companies chapter into the 2013 Act itself, brought into force in February 2021, at which point the last substantial compartment of the 1956 Act still in daily operation was finally closed, eight years after the repeal clause had been enacted.

One compartment of the new Act itself repeated the old Act’s central mistake, and its fate completes the picture. The Companies Act 2013 as enacted contained a chapter on the revival and rehabilitation of sick companies, Sections 253 to 269, which was to be the third attempt, after the sickness statute of 1985 and the unnotified code of 2002, to give sick industrial companies a statutory rescue process. The chapter was never notified. When the Insolvency and Bankruptcy Code was enacted in 2016, its schedule of amendments omitted the chapter from the Companies Act altogether, uncommenced and unlamented, and the subject passed to the Code’s resolution process. A provision can therefore die at three removes: enacted in a replacement statute, never brought into force, and deleted by a later statute before it ever operated. The episode is a useful discipline against triumphalism about 2013. The replacement method solved the problem of the old Act’s text. It did not solve the underlying problem, which is that Indian commencement practice leaves the real existence of enacted law to executive notification, and a provision dependent on an institution that has not been built, or on a policy that has changed since enactment, may wait indefinitely in the Gazette. The 1956 Act’s tribunal waited fourteen years. The 2013 Act’s sickness chapter waited three years and was then removed. The difference is that the second failure was tidied away quickly, which is progress of a sort.

The practical consequences of this method deserve plain statement, because they are the part of the replacement that practitioners actually lived. For a period of years, a single company could be simultaneously a 2013 Act company for its board meetings and accounts, a 1956 Act litigant in an oppression petition filed before the transfer, and a stranger to both in a winding up that had begun in a High Court in the previous decade. Compliance calendars had to be rebuilt twice, once in September 2013 and again in April 2014, on the strength of circulars that arrived after the notifications they explained. The ministry issued a long series of clarifications and removal-of-difficulties orders, a form of executive patching that the removal of difficulty power in the new Act contemplated for its first five years. None of this was chaos, exactly; the savings architecture held, and no proceeding fell into a void. But the transition imposed a real tax in uncertainty, and it generated the professional complaint, recorded across the commentary of the period, that India had replaced the uncertainty of an obsolete statute with the uncertainty of two statutes at once. Anyone advising on a live matter from those years will recognise the further truth that the general account in an article of this kind cannot substitute for advice on the individual case, because which regime governed turned on the stage the proceeding had reached and the date on which each relevant provision had been notified, and those are questions of record, not of impression. For the reader keeping the documentary trail of that transition straight, working through scanned circulars, commencement notifications and legacy registry extracts, an office file viewer for opening the legacy document formats is the practical companion to the analysis.

The dual-regime years as they were lived

The transition deserves one more pass at ground level, because the lived experience of the dual-regime years is the strongest evidence on both sides of the argument about how replacement should be executed. Consider what the staged method asked of an ordinary company secretary in the eighteen months after September 2013. The definitions had changed beneath her: the new Act’s definition of a private company altered the ceiling on members and carried different consequences, and the ministry had to clarify that memoranda received before the notification date would be registered under the old definition. The financial year was unified by the new Act to end on 31 March for all companies, a rational reform that nonetheless forced every company with a different year-end through an alignment transition. Board reports, auditors’ reports and annual returns acquired new contents in April 2014 under rules notified weeks before they took effect, in some cases days. The acceptance of deposits, an area where the old Act’s regime had its own mature case law, was re-founded on new provisions whose interaction with deposits already accepted under the old law required a sequence of clarifying circulars. Each of these was manageable. Their accumulation was the cost.

Set against that cost is the counterfactual, which the defenders of staging were entitled to press. A single commencement of all 470 sections on one date would have required the rules, the forms, the electronic filing system, the tribunal and the audit authority to be ready simultaneously, and the tribunal at least could not have been ready, because its design was still before the Supreme Court until 2015. The choice was never between order and disorder. It was between a disorder of sequence, in which the law changed in tranches and practitioners complained about uncertainty, and a disorder of simultaneity, in which the law changed at once and some of its institutions did not exist. The Indian state’s wider practice in the same decade points the same way: when the criminal law codes were replaced a decade later, the transition again proceeded by appointed dates with savings for pending proceedings, and again attracted the professional complaint of dual-regime uncertainty that this cluster records for company law. Staging, in other words, is not an accident of this replacement but the standard Indian technology for replacing a working code, and its costs are the recurring price of the method. Whether the price was minimised here is arguable; that some price was unavoidable is not.

The transition worked through: three files

Abstract savings language becomes concrete when it is run against files that were actually open during the transition, and three stylised files, each typical of thousands, show what the machinery did.

Take first a prosecution. In 2012 a Registrar filed a complaint against a company and its officers for a default committed under the 1956 Act, and the case was pending before a magistrate when the repeal took effect. The repeal did not touch it. The savings clause provides expressly that a prosecution instituted under the repealed enactments and pending before a court continues to be heard and disposed of by that court, and the General Clauses Act savings behind it preserve the liability incurred under the old law. The officers’ exposure was therefore measured by the old Act’s penalties for the old Act’s offence, years after the Act itself had been repealed, which is why reports of prosecutions under the Companies Act 1956 continued to appear in the legal press long after 2014 without any error being involved.

Take next an oppression petition. Members of a private company, locked in a dispute with its controlling family, petitioned the Company Law Board in 2015 alleging oppression and mismanagement. They petitioned the Board, rather than the Tribunal the new Act promised, because the savings proviso kept the Board’s jurisdiction under the old Act alive and the new Act’s corresponding provisions had not been notified. In June 2016 the Board was dissolved and the Tribunal constituted, and the pending petition transferred to the Tribunal with the rest of the Board’s docket, to be decided under the law that had governed it when filed. The litigants’ forum changed mid-dispute by operation of law, their pleadings did not, and the substantive test the Tribunal applied was the one the parties had framed their case around in the old institution. Multiply that file by the Board’s whole pending caseload and the scale of the 2016 transfer, the moment the repeal’s longest proviso finally spent itself, becomes visible.

Take finally a scheme of arrangement. A company that had obtained a High Court’s directions for meetings of its creditors and members in 2015, under the old Act’s compromise provisions as preserved in the company court, found the forum for the sanction stage reassigned as the compromise provisions of the new Act were notified in December 2016 and pending proceedings were transferred to the Tribunal in the stages the transfer notification prescribed, with proceedings that had reached specified stages remaining with the court. The commercial transaction had to be steered, by its advisers, across a change of forum, a change of procedural rules and a change of governing text in mid-course. That the system absorbed these transfers without a reported breakdown is a credit to the savings drafting. That it imposed them at all is the cost side of the ledger, and both entries belong in an honest account of what replacing a working code involves.

The executive as repair shop: removal of difficulties

A word should be said about the legal instrument that held the transition together at its roughest points, because it too is part of the replacement’s design and part of its lesson. The new Act contained the standard power, in Section 470, for the Central Government to make orders, consistent with the Act, to remove difficulties arising in giving effect to it, exercisable during the first five years. The transition used the power repeatedly, and used the ministry’s general circulars even more freely, to answer the questions the staged commencement generated: which definition governed a filing straddling a notification date, how resolutions passed under one regime counted under the other, how deposits accepted under the old law were to be treated. Each answer was sensible in itself. Their cumulative effect was that a material part of Indian company law during the transition years existed not in either statute but in the ministry’s clarifications upon them, instruments that are not laid before Parliament, are not always easy to find after the event, and bind in practice through the department’s administration rather than through adjudication. Practitioners of the period will remember the discipline the years imposed: no question about the transition could be answered from the two Acts alone, and the file on any live matter had to carry the circulars with the sections. A replacement strategy that reserves its hardest mapping questions to executive clarification is choosing administrative flexibility over legal certainty, and choosing it at the moment of maximum confusion. The choice was defensible, for the alternative was a commencement schedule of impossible completeness drafted in advance. But it should be seen clearly for what it was, because it recurs in every Indian code replacement and because its costs fall, invisibly and regressively, on the smallest companies least able to buy the advice that navigates it.

What the replacement kept, discarded and added

A replacement is three operations, not one, and the 2013 Act performed all three, though a reader who wants the detailed ledger of differences should go to the article that compares the Companies Act 1956 and the Companies Act 2013 provision by provision; what follows is the shape of the three operations as the replacement executed them.

What was kept deserves a further word at the level of the individual company, because continuity was experienced not as doctrine but as paperwork that did or did not have to be redone. A company’s certificate of incorporation under the old Act remained its certificate; no re-registration was required, and the savings clause deems the company to be governed by the new Act as if registered under it. The director identification numbers allotted under the 2006 machinery continued as the same numbers in the new system. Charges registered against a company’s assets under the old Act stood registered, and the registers and funds of the old regime were deemed to be registers and funds under the corresponding provisions of the new. Companies that held licences under the old Act’s Section 25, the provision for associations with charitable and like objects formed as limited companies, continued under the new Act’s corresponding licence regime in Section 8, their licences intact. Even the memorandum and articles, drafted against the old Act’s model tables, continued in force, subject to the new Act’s overriding provisions where the two conflicted, which the new Act states expressly. The design intention is unmistakable and correct: the replacement was of the law, not of the companies living under it, and every device in the savings architecture points the same way. The million-plus files in the registry were carried across whole, and the disruption of the transition was concentrated where it could not be avoided, in the procedures and obligations of the years in which the change was executed.

What was kept was the skeleton. Incorporation by registration, the memorandum and articles as the constitutional documents of the company, limited liability, the Registrar’s registry, the annual cycle of accounts, audit and general meeting, the distinction between private and public companies, and the state’s powers of investigation all passed from the old Act to the new, because they were the working machinery of a system with well over a million companies on the register and no one proposed to re-found it. Continuity of incorporation was guaranteed expressly by the savings clause, and the great body of case law built on the old Act’s concepts, on meetings, on directors’ duties at common law, on the construction of articles, continued to be cited under the new Act wherever the successor text reproduced the substance of the old.

The discarded category can be specified a little further without crossing into the companion article’s ledger, because the categories of discard are themselves the argument. The replacement discarded the approval as the default instrument of control: the old Act’s habit of requiring the government’s prior sanction for the ordinary decisions of boards, appointments, remuneration, loans, contracts with interested directors, was replaced as a matter of design by disclosure, shareholder approval and ex post liability, the philosophy the Irani report had announced. It discarded the statute as the repository of forms and fees, moving that stratum wholesale into rules. It discarded the assumption that one undifferentiated code could serve the one person enterprise and the listed conglomerate, introducing calibrated categories instead. And it discarded, perhaps most consequentially for daily practice, the department as adjudicator of first resort: the quasi-judicial work the Company Law Board had done for the old Act was assigned to a tribunal outside the ministry, so that the successor statute’s disputes would be heard by a body the administering department did not staff. Each of these discards is a change of constitutional texture within company law, a redistribution of power among the state, the board, the shareholders and the courts, and none of them was achievable by the amendment of individual sections, because each was a property of the old Act’s whole arrangement rather than of any provision in it. That, in the end, is the technical core of the case for replacement, and it is stronger than the rhetoric of age and length that usually carries the argument in public.

What was discarded was, first, bulk: hundreds of provisions that had accumulated to deal with situations the market had outgrown, the managing agency’s long afterlife in the text, licensing-era approval requirements that the reform decades had already hollowed out, and procedural detail that the Irani design reassigned to rules. Discarded too was the old Act’s terminal machinery as a going system: the sickness jurisdiction was never revived in the new Act and its chapter was deleted unborn, and court-conducted winding up gave way to the Code. And discarded, deliberately, was the assumption that adjudication belonged in the department and the generalist courts: the new Act’s centre of gravity for disputes is the Tribunal, an institution the old Act had promised and failed to deliver.

What was added is the part of the replacement that reads as a direct reply to the failures catalogued above. An independent audit regulator, the National Financial Reporting Authority, created by the new Act though constituted only in 2018, answered the oversight gap the Naresh Chandra Committee had identified fifteen years earlier. A class action provision answered the minority remedy gap, at least in text. The Serious Fraud Investigation Office, which had existed as an executive office since 2003, was given statutory standing and powers. Independent directors were defined and given a code, auditors were subjected to rotation and to a bar on non-audit services, related party transactions were brought under a single disciplined provision, a mandatory corporate social responsibility expenditure was imposed on large companies, and new company forms, the one person company and the small company, recognised the scale diversity the old Act had ignored. One addition did not come from the committees at all, and its route into the Act is a reminder that a replacement bill, once in Parliament, is a vehicle onto which the politics of the moment can load cargo the designers never contemplated. The mandatory corporate social responsibility obligation, requiring companies above prescribed thresholds to spend two per cent of average net profits on scheduled social purposes or explain the failure, has no counterpart in the 1956 Act and was not part of the Irani Committee’s design in its mandatory form; the committee had contemplated social responsibility in the register of encouragement and disclosure. The obligation in its enacted form took shape during the bills’ long passage, through the standing committee examinations and the government’s amendments, and entered the statute as a distinctively Indian experiment that no other major company law imposed in that form. Its presence in the final Act matters for the history in two ways. It demonstrates that the nine years were not merely delay: the text that emerged in 2013 was not the text designed in 2005, and some of the difference is the deposit of the passage process itself, for good and ill. And it complicates the missed-opportunity criticism considered below, because a Parliament that used the replacement to enact a new social obligation on companies was not simply failing to follow a blueprint; it was legislating, which is what Parliaments are for, and the blueprint’s authors never had a monopoly on the purposes a new companies Act might serve.

Whether each addition has worked is a question for the implementation record and for the pillar article; the point here is that the additions were not decorative. Each maps to a named failure of the predecessor, which is what distinguishes a replacement argued from a diagnosis from a mere change of fashion.

The register the Act left behind

One inheritance of the old Act deserves separate mention because it was measured in entities rather than doctrines. The Act made incorporation easy and exit hard, and six decades of that asymmetry produced a register carrying a very large population of companies that had ceased to trade but had never been wound up or struck off: companies whose directors had walked away, whose registered offices were shuttered, whose filings had stopped years before. The old Act contained a strike-off power for defunct companies, but it was hedged, slow and little used relative to the accumulation, and the official liquidators and registrars charged with the register’s hygiene were the same understaffed offices that handled everything else. The register’s unreliability was not a clerical footnote. Creditors, investors and counterparties read the register as a statement about who exists and who stands behind a name, and a register crowded with the dead misstates the corporate population of the country and shelters the shell company, the entity kept for its registration rather than its business.

The successor regime inherited the population along with the register. The new Act carried a strike-off provision of its own and added the concept of the dormant company, a formal status for the inactive-but-honest entity, which was the Irani design’s civilised answer to the problem: give the genuinely inactive company a lawful way to sleep, so that enforcement can concentrate on the ones that are neither trading nor sleeping. In the years after the main commencement the ministry conducted strike-off drives that removed defunct companies from the register in very large numbers, an administrative campaign whose scale was itself a measurement of what the old Act had accumulated. The episode closes the circle of this article’s argument. Even the register, the most basic artefact of company law, could not be cleaned within the old Act’s working methods, because those methods were the cause of the dirt. It took a new statute, a new status category and a new administrative campaign to do what sixty years of amendment had not attempted.

Continuity across the repeal: what a savings clause cannot do

The savings clause carried legal positions across the repeal; it could not carry everything, and the residue is worth examining because it is where replacement imposed its least visible costs. A savings clause preserves acts done, rights accrued and proceedings pending. It does not preserve the interpretive world in which a statute lived: the departmental circulars that told registrars how a provision was administered, the settled practices of company courts on matters of discretion, the professional folklore about which filings were scrutinised and which were waved through. Some of that world transferred naturally, because the successor text reproduced the old provisions and the courts applied the settled rule that where a provision is re-enacted in the same terms after being judicially construed, the construction is presumed to have been adopted with it. Some of it did not transfer, because the new Act changed the forum: the discretionary case management of oppression petitions, built up over decades in the Company Law Board’s practice, had to be rebuilt in a tribunal that began with no accumulated practice of its own.

The human transition was managed by a provision easily overlooked. Section 466 of the new Act dealt with the dissolution of the Company Law Board and carried its chairperson, members, officers and staff across to the Tribunal, so that the institution that had administered the old Act’s busiest remedy supplied the first personnel of the institution that replaced it. The last chairperson of the Board became the first president of the Tribunal. There is a symbolism in that arrangement which the drafters may or may not have intended: the replacement of Indian company law was executed substantially by the people who had operated the law being replaced, applying a new text to problems they already knew. It is one reason the transition, for all its documentary confusion, never produced an adjudicatory vacuum.

The deeper continuity is doctrinal, and it qualifies any dramatic telling of the replacement. Directors’ duties in Indian law were, under the old Act, a compound of the articles, the statute’s specific prohibitions and the common law; the new Act codified them, but the codification was read through the case law the old regime had produced. The doctrines governing meetings, resolutions, the rule in Foss v. Harbottle and its exceptions, the construction of the memorandum as the company’s charter, the protection of outsiders dealing with the company: all of this crossed the repeal intact, because it had never lived in the repealed text alone. A repeal in a common law system kills a text, not a jurisprudence. The practitioner of 2015 citing a 1975 authority on the indoor management rule was not defying the repeal; the repeal had been drafted, through its savings and through the successor’s deliberate re-enactments, to leave that body of law standing. Understanding this is essential to a fair assessment of the replacement’s scale. What changed in 2013 and after was the operating system of Indian company law: its forums, its regulator, its sanctions and its compliance machinery. The conceptual grammar of the subject changed far less, and the claim sometimes heard, that India acquired a new company law in 2013, is true only in the first of those senses.

Why replacement beat amendment: the table

The argument of this article, reduced to the artifact a reader can keep, is the following table. Each row is a failing of the 1956 Act that the reform record identified; the middle column gives the honest answer to whether one more amendment could have fixed it; the last column records what the replacement did instead.

Failing of the 1956 Act	Could amendment have fixed it?	What the 2013 Act did instead
Length and structure: 658 sections in thirteen parts, obligations written into the statute in fine detail	In principle yes, but only by an amendment so large it would have been a new Act wearing an amendment’s name	Recast the law as 470 sections in twenty-nine chapters and moved operative detail into rules, as the Irani Committee recommended
Volume of amendment: twenty-four amendments that left the text a palimpsest of incompatible eras	No. Amendment was the cause of the condition, not a cure available for it	Repealed and restated the whole law in one text, with Section 465 savings to carry accrued rights across
Obsolete winding-up machinery: liquidation by the High Courts over years and decades, assets decaying in process	Partly. Procedure could be tightened, but the forum was the defect, and moving the forum meant new institutions	Designed tribunal winding up, then let the Insolvency and Bankruptcy Code 2016 take the subject over with a timed process
No independent audit regulator: discipline left to the profession’s own institute	Yes in theory; a regulator could have been inserted, as other regulators were created by standalone statutes	Created the National Financial Reporting Authority in the Act itself, constituted in 2018
Weak minority remedies: oppression petitions for the few, no class action for investors as a group	Yes in theory; new remedies can be added by amendment	Added a class action provision and restated oppression and mismanagement before the Tribunal
The unconstituted tribunal: created in 2002, held defectively designed in 2010, never notified	Demonstrated no. The amendment existed for fourteen years and produced no institution	Rebuilt the tribunal provisions in the new Act and constituted the Tribunal on 1 June 2016

The honest assessment

Two verdicts on the replacement compete, and the stance of this series is to give each its strongest form before judging between them.

The defence runs as follows, and it was made at the time by the government that carried the bill: the Minister of State for Corporate Affairs, Sachin Pilot, told the Rajya Sabha during the passage debate that the new law would bring India’s corporate governance into line with a changed global business environment and would serve for decades. On the structural questions the defence is strong. The replacement did what amendment had repeatedly failed to do: it produced a single coherent text, it created the tribunal, it built the audit regulator into the statute, it gave the fraud office statutory powers, and it moved the detail of compliance into rules that can be revised without a parliamentary session. The staged commencement, for all its friction, avoided the alternative of a single commencement date for 470 sections and the chaos of a registry, a profession and a judiciary asked to change every practice on one morning. And the savings architecture worked: there is no reported instance of proceedings voided or rights extinguished by the repeal itself, which is the first test a repeal clause has to pass.

The case for the other route, which was also made and deserves its strongest form, is the case for continuity. Company law is infrastructure: over a million companies, their bankers, their auditors and their shareholders had organised their documents and their expectations under the 1956 Act, and sixty years of case law gave its provisions settled meanings that practitioners could price. A replacement, however carefully saved, disturbs that settlement everywhere at once, invalidates a profession’s accumulated fluency, and replaces known defects with unknown ones. Representations made to the Standing Committee during the bill’s examinations pressed versions of this argument: that the transition should be slower, the obligations phased, the smaller companies shielded, and the commencement made conditional on the rules and the tribunal being ready. Parts of that case were accepted, and the staged commencement itself is a concession to it. But the continuity argument proves too much if it is pressed to its conclusion, because it applies with equal force against every replacement of every mature code, and its practical demand, indefinite amendment of a text whose amendment mechanism had itself failed, was the one option the record had already refuted. Continuity is a value in the execution of a replacement. The committees were right to treat it as a weak reason against replacement itself.

The criticism, pressed by practitioners and by commentators who had supported the Irani blueprint, is that the replacement was a missed opportunity of a specific kind. The Irani Committee had proposed a compact, principles-based statute. What emerged was an Act of 470 sections which, with its rules, forms and circulars, imposed a compliance load at least as heavy as the old regime, and which in its first years required a stream of exemptions, clarifications and amendments, including a comprehensive amendment within two years of the main commencement, to make it workable. On this view India kept the 1956 Act’s regulatory instincts, its suspicion of the board and its appetite for prescribing detail, while changing the binding. The criticism has force, and the early amendment history of the 2013 Act supports it. Within two years of the main commencement the new Act required a comprehensive amendment of its own, the Companies (Amendment) Act 2015, which among other corrections removed the minimum paid-up capital requirement that the new Act had imposed on incorporation, made the common seal optional, and softened provisions that had proved unworkable in their first year. The government then appointed a Company Law Committee in 2015 to review the Act’s implementation, and its report of February 2016 recommended changes running across the statute, which fed the amending Acts of the following years, including the instalments of decriminalisation of procedural defaults that the committees of the old regime had demanded in vain. Read one way, this record shows the replacement reproducing the disease of the predecessor: an Act under permanent amendment from birth. Read more carefully, it shows something different and more hopeful. The amendments of the 2013 Act were, in the main, corrections at the margin of a structure that held: they did not need to create its institutions, supply its missing remedies or rescue its uncommenced chapters, because those things existed. The 1956 Act’s amendments were attempts to make an obsolete design do new work. The 2013 Act’s amendments were the ordinary maintenance of a working one. The distinction is the difference between renovation and repair, and it is the practical test by which the replacement decision should be judged. But it should be aimed precisely. The failure it identifies is not a failure of the replacement decision; it is a failure to follow the replacement’s own blueprint. Nothing in the criticism suggests that a twenty-fifth amendment of the 1956 Act would have produced a lighter statute, and the one experiment that tests the counterfactual, the fate of the 2002 amendment, suggests the opposite: the old frame absorbed reforms and neutralised them.

The principle the episode establishes

This cluster is where the series’ recurring inquiry into notification and commencement meets corporate law, and the episode states a general principle with unusual clarity. A legislative fix that depends on a new institution is only as good as the state’s willingness and ability to build the institution. The 1956 Act’s last decades are a catalogue of the principle’s violations: a tribunal enacted and unbuilt for fourteen years, a sickness code enacted and never notified, a repeal enacted and left uncommenced for thirteen years, an audit regulator recommended in 2002, legislated in 2013 and constituted in 2018. In each case the text was not the constraint. Parliament had spoken, in some cases repeatedly. The constraint was the sequence of unglamorous acts, designing appointments that can survive constitutional scrutiny, constituting selection committees, notifying commencements in the right order, staffing registries, that turns an enactment into a fact. Replacement succeeded where amendment failed not because its text was cleverer, though in places it was, but because the replacement project eventually carried its institutions with it: the tribunal was constituted, the fraud office was given its powers, the audit authority was, after a further wait, set up. Where the replacement repeated the old pattern, as with its sickness chapter, it failed in the old way, and was quietly overtaken by a code that arrived with its institutions attached.

For the legislator or staffer who is this article’s first reader, the transferable lesson concerns the drafting of commencement and savings provisions, which are usually treated as boilerplate and were, in this replacement, the operative law for years. Section 465’s provisos, keeping the old adjudicatory jurisdiction alive until the transfer date and keeping the producer companies chapter alive until its successor existed, are what stand between a staged replacement and a legal vacuum, and the corresponding-provisions confusion of September 2013 is what happens when the mapping between old and new is left to circulars after the event. A replacement bill that names, in its own text, the fate of each compartment of the old law on each commencement date is harder to draft and incomparably easier to live under. The 1956 Act’s replacement got that architecture half right in the statute and improvised the other half in the ministry, and the division between the two halves is precisely the division between the parts of the transition that worked silently and the parts that generated years of professional complaint.

The verdict this article defends is therefore the narrow one the record supports. Replacement was justified, and justified by the strongest piece of evidence in the whole story: a statute whose Parliament-approved tribunal could not be constituted for fourteen years had lost the capacity to be repaired by the ordinary means of repair. The nine years, the three bills and the staged repeal were not legislative vanity. They were the cost of doing in one statute what the Indian state had been unable to do by instalments since 1993. Whether the successor statute justified that cost in its day-to-day operation is the question its own cluster of articles takes up; the question of why its predecessor ended has, on the record set out here, a clear answer. The counterfactual can be stated fairly, and stating it shows why the verdict survives it. Suppose the replacement bills had failed a third time and India had entered the following decade still amending the Act of 1956. The tribunal would still have required constitution, and would have required it under a text the Supreme Court had held defective, so a corrective amendment would have been the first necessity, an amendment whose subject was the repair of a previous amendment. The audit regulator would have required creation, achievable by amendment or by standalone statute, against the same professional resistance that had stalled it since 2002. The class action, the fraud office’s statutory powers, the recasting of detail into rules: each was individually enactable, and the list is precisely the list of things that amendment cycles had been failing to enact, one by one, for twenty years, because in an amending bill each item fights alone against its own opponents, while in a replacement the items carry one another. The strongest argument for the replacement method was never that amendment was impossible in logic. It was that amendment had a twenty-year record in fact, and the record was the 2002 tribunal, the 1993 and 1997 bills in their graves, and a sickness code that had never spent a day in force. Legislatures, like courts, are entitled to decide on the evidence of what has actually happened when the alternative was tried.

The Companies Act 1956 ended because amendment had been tried, at length and in good faith, and the tribunal that never sat is the proof.

Frequently Asked Questions

Q: Why was the Companies Act 1956 replaced?

It was replaced because amendment had demonstrably stopped working. The Act had been amended twenty-four times, yet its central defects were structural rather than textual: obligations written into the statute in fine detail so that every change in practice required Parliament, adjudication scattered among High Courts and a departmental board, no independent audit regulator, no class action for investors, and a winding-up process that consumed years. The decisive proof was its own 2002 amendment, which created a National Company Law Tribunal that could not be constituted for fourteen years because its design of appointments and qualifications was held constitutionally defective. When a statute’s most recent major repair cannot be brought into force, replacement rather than a twenty-fifth amendment becomes the rational course.

Q: What did the JJ Irani Committee recommend?

The Expert Committee on Company Law chaired by Dr J. J. Irani, appointed in December 2004 and reporting on 31 May 2005, recommended that India enact a new, compact and principles-based companies statute rather than amend the 1956 Act again. The Act itself should state governing principles and hard obligations, while forms, thresholds, procedures and timelines should move into rules that the executive could revise without a parliamentary session. It proposed a single governance framework calibrated to the size and public character of companies, recognition of the one person company, stronger board and disclosure discipline, independent oversight of auditors, a statutory serious fraud investigation office, better minority protection, easier exit for defunct companies, and the single tribunal the Eradi Committee had proposed for company disputes and winding up.

Q: Did the Satyam scandal cause the new company law?

No, though it changed the new law’s politics and sharpened some of its provisions. The reform process predated the scandal by years: the Eradi Committee reported in 2000, the Naresh Chandra Committee in 2002, the Irani Committee designed the replacement in 2005, and the first Companies Bill embodying that design was introduced in October 2008. Satyam’s chairman confessed the falsification of the accounts, amounting to Rs 7,136 crore, on 7 January 2009, after that bill had lapsed with the dissolution of the Lok Sabha. What the scandal did was make the case for an independent audit regulator, a statutory fraud office and stronger fraud provisions almost unanswerable in committee, and raise the bill’s priority in a crowded legislative programme.

Q: How was the Companies Act 1956 structured?

As enacted, the Companies Act 1956 ran to 658 sections arranged in thirteen parts, with fifteen schedules, making it one of the longest statutes on the Indian book. The early parts dealt with incorporation, the memorandum and articles, prospectuses and share capital. The long central part governed management and administration: directors, meetings, accounts, audit, dividends and the registers and returns filed with the Registrar of Companies. Later parts covered compromises and reconstructions, the prevention of oppression and mismanagement, government companies, and finally winding up, which occupied a substantial share of the statute and was conducted through the High Courts. Around the text stood its administrative machinery of Registrars, regional directors and, from the early 1990s, the Company Law Board.

Q: Why could the Companies Act 1956 tribunal provisions not be brought into force?

The Companies (Second Amendment) Act 2002 inserted provisions for a National Company Law Tribunal and an appellate tribunal into the 1956 Act, but commencement depended on notification, and the design was challenged first. The Madras High Court held key provisions defective in 2004, and a Constitution Bench of the Supreme Court, in Union of India v. R. Gandhi in May 2010, upheld Parliament’s power to create the tribunals and transfer company jurisdiction to them while holding the provisions on qualifications and appointment of members defective: they admitted persons of insufficient seniority to do the work of High Court judges and gave the executive too large a role in selection. The corrections the Court required were never made to the 1956 Act, so the tribunal was never constituted under it. It was finally constituted on 1 June 2016 under the Companies Act 2013.

Q: What happened to cases pending under the Companies Act 1956?

They were carried across by the savings machinery rather than extinguished by the repeal. Section 465 of the Companies Act 2013 preserves things done, proceedings taken and penalties imposed under the repealed Act, provides that prosecutions pending before a court continue to be heard by that court, that inspections and investigations ordered under the old Act continue as if ordered under the new, and that matters pending before the Registrar, a regional director or the Central Government are concluded under the old Act despite its repeal. Cases before the Company Law Board, whose jurisdiction the savings proviso expressly kept alive, transferred to the National Company Law Tribunal when the Board was dissolved on 1 June 2016, and winding-up proceedings in the High Courts transferred in stages from December 2016, with the most advanced staying in court.

Q: How many times was the Companies Act 1956 amended?

The count most often recorded, including by the National Foundation for Corporate Governance, is twenty-four amendments between 1956 and the final amendment in 2006. Some were minor corrections, but several were structural: the tightening amendments of the 1960s, the large overhaul of 1988 following the Sachar Committee, the governance amendments of 2000 that introduced audit committees and the postal ballot, the two amendments of 2002 that created producer companies and the stillborn tribunal, and the amendment of 2006 that introduced director identification numbers and electronic filing. The count itself became an argument in the replacement debate, since a statute amended two dozen times had become a palimpsest of incompatible regulatory eras that no reader could take in as a whole.

Q: Was the Companies Act 1956 too long to administer?

Length alone was not the defect, but length of a particular kind was. The Act’s 658 sections wrote operative detail, thresholds, procedures and forms, into primary legislation, so the law could not move until Parliament moved; they accumulated across twenty-four amendments into a text whose organisation recorded the order of its accidents rather than the logic of its subject; and they fell on nearly every company regardless of size, multiplying technical defaults. The Irani Committee’s terms of reference included compactness and the removal of redundant provisions, and its answer was not merely a shorter Act but a different distribution of law: principles and hard obligations in the statute, changing detail in rules. The successor Act’s 470 sections, with much of the operative machinery in rules beneath them, embody that answer.

Q: How long did the replacement of the Companies Act 1956 take?

Measured as a legislative project, about nine years: the Irani Committee was appointed in December 2004, and the Companies Act 2013 received presidential assent on 29 August 2013 after three bills, those of 2008, 2009 and 2011, and two examinations by the Standing Committee on Finance. Measured as a transition on the ground, it took much longer. The new Act commenced in tranches, ninety-eight sections in September 2013 and a further 183 in April 2014; the Tribunal was constituted only in June 2016; winding up moved to the new insolvency regime from December 2016; the audit authority was constituted in 2018; and the producer companies chapter of the repealed Act remained in operation until February 2021. The last compartments of the replacement were still closing eight years after the repeal clause was enacted.

Q: Which parts of the Companies Act 1956 survived its repeal, and for how long?

Section 465 itself kept two compartments alive expressly. The provisions governing the jurisdiction of the Company Law Board and the courts continued to apply as if the Act had not been repealed until the transfer of matters to the Tribunal, which occurred when the Tribunal was constituted on 1 June 2016. Part IXA, the producer companies chapter, continued to apply to producer companies until a successor regime existed, which arrived only when a producer companies chapter was written into the Companies Act 2013 and brought into force in February 2021. Beyond those express survivals, the general savings preserved accrued rights, liabilities, registrations and pending proceedings indefinitely, and provisions of the old Act borrowed by the Limited Liability Partnership Act 2008 continued for partnerships until replaced by corresponding provisions of the new Act.

Q: What did Section 465 of the Companies Act 2013 save?

Section 465 is the repeal and savings clause, and its savings are enumerated rather than general. Anything done under the repealed Acts, including rules, orders, appointments, inspections, proceedings and penalties, is deemed done under the corresponding provisions of the new Act so far as consistent with it. Companies incorporated under the old Act remain validly incorporated and are governed by the new Act. Registers and funds continue. Prosecutions pending in court continue before that court. Inspections and investigations ordered under the 1956 Act continue as if ordered under the corresponding provisions of the 2013 Act. Matters pending before the Registrar, a regional director or the Central Government are concluded under the old Act notwithstanding repeal. The section also preserves the general operation of Section 6 of the General Clauses Act 1897 on the effect of repeal.

Q: What role did the Bhabha Committee play in the Companies Act 1956?

The Company Law Committee chaired by Homi J. Bhabha, appointed in 1950 and reporting in 1952, was the drafting basis of the 1956 Act. It examined the working of the Indian Companies Act 1913 and the English Companies Act 1948 and recommended a distinctively Indian settlement: fuller disclosure in prospectuses and accounts, tighter control over managerial personnel and the managing agency system, stronger governmental powers of investigation and intervention, and protection for minority shareholders against controlling groups. Parliament enacted that settlement, with modifications, as Act 1 of 1956. The committee’s role is the standing answer to the claim that the 1956 Act was a colonial inheritance: its text was written after independence, by an Indian committee, for the economy of the new republic.

Q: What happened to companies registered under the Companies Act 1956?

Nothing happened to their existence. The savings clause in Section 465 provides that the incorporation of companies registered under the repealed enactments continues to be valid and that the provisions of the new Act apply to them as if they were registered under it. No company had to re-incorporate, re-register or surrender its certificate when the repeal took effect; its memorandum and articles continued in force subject to the new Act’s overriding provisions, its director identification numbers continued, and charges registered against its assets stood. What changed was the law governing its future conduct: from the commencement dates of the corresponding provisions of the Companies Act 2013, its board meetings, accounts, audit, filings and disputes were governed by the new statute rather than the old.

Q: Why did the Companies Bills of 2008 and 2009 never become law?

The Companies Bill 2008, introduced in the Lok Sabha on 23 October 2008, lapsed when the fourteenth Lok Sabha was dissolved ahead of the 2009 general election, as bills pending in the House do on dissolution. The Companies Bill 2009, introduced on 3 August 2009, did not lapse but was overtaken by its own scrutiny: the Standing Committee on Finance, reporting on 31 August 2010, recommended changes extensive enough that the government chose to withdraw and recast the bill rather than amend it in passage. The recast text became the Companies Bill 2011, which the same committee examined again before its passage in December 2012 and August 2013. The two failures are of different kinds, one mechanical and one deliberative, and both contributed to the replacement’s nine-year gestation.

Q: Did the Companies Act 1956 extend to the whole of India?

Not as enacted. Section 1 of the Act provided that it extended to the whole of India except the State of Jammu and Kashmir, which stood outside it under the constitutional arrangements of the time and had company legislation of its own. Sikkim presented a second pocket: it retained its own incorporation statute, the Registration of Companies (Sikkim) Act 1961, even after its merger with India, because the Companies Act 1956 was not extended to it in the ordinary way. Section 465 of the Companies Act 2013 repealed the Sikkim Act along with the 1956 Act, so the replacement of 2013 was also the moment at which a single incorporation code first covered the whole territory in force under the new Act’s own extent provision.

Q: Why did producer companies stay under the repealed 1956 Act until 2021?

Because the Companies Act 2013 as enacted contained no producer companies regime to succeed to them, and Parliament chose continuity over a vacuum. Producer companies, the farmer-producer form inserted into the 1956 Act as Part IXA by the amendment of 2002, were therefore expressly carved out of the repeal: Section 465 provided that Part IXA would continue to apply to a producer company as if the 1956 Act had not been repealed until a special law was enacted for producer companies. New producer companies continued to be incorporated under the chapter of a repealed statute for nearly eight years. The successor arrived when the Companies (Amendment) Act 2020 inserted a producer companies chapter into the 2013 Act, brought into force in February 2021, and the proviso could finally be spent.

Q: What did the Companies (Amendment) Act 2000 change?

The amendment of 2000 was the principal attempt to write the corporate governance agenda into the 1956 Act rather than leave it to the securities regulator’s listing agreement. It required public companies above a prescribed paid-up capital to constitute audit committees of the board, formalised the directors’ responsibility statement in the board’s report, introduced the postal ballot so that specified business could be transacted by shareholders voting by post, and revised the provisions on buy-back of shares, debentures and the nomination facility for shares and deposits. Its significance in the replacement story is that it demonstrated both the possibility and the limit of the amendment method: the governance content could be added, but it arrived as one more stratum on an unreformed structure, duplicating in the statute what Clause 49 of the listing agreement already imposed on listed companies.

Q: When did company filings in India move to electronic filing?

The statutory basis was laid by the Companies (Amendment) Act 2006, which inserted provisions for the filing of documents in electronic form and for director identification numbers, and the machinery was the MCA21 e-governance programme of the Ministry of Company Affairs, which moved the registry from paper files held in state Registrar offices to a single electronic system during 2006. Electronic filing therefore arrived seven years before the Act itself was replaced, and through an amendment, which is worth noting against any claim that amendment could achieve nothing: the 2006 amendment changed the medium of administration successfully and permanently. What it could not do was change the substance being administered, and the registry’s new electronic front office spent its first years processing the obligations of a statute designed in 1956.

Q: Was the Companies Act 1956 a colonial law?

No. The claim is widespread and wrong. The 1956 Act was enacted by the Parliament of independent India as Act 1 of 1956, on the basis of the report of the Company Law Committee chaired by Homi J. Bhabha, an Indian committee appointed in 1950. Its ancestry was partly English, as the ancestry of most Indian codes is: it consolidated and replaced the Indian Companies Act 1913, which had followed English models, and the Bhabha Committee worked with the English Companies Act 1948 in view. But ancestry is not authorship. The 1956 Act’s supervisory philosophy, its controls on managing agencies and managerial remuneration, its provisions for government companies and its investigation machinery were Indian choices made for the Indian economy of the 1950s, and several of them departed deliberately from the English pattern of the day.

Q: Who supervised auditors under the Companies Act 1956?

In substance, the profession supervised itself. Auditors were appointed by shareholders in general meeting and reported to them, but discipline for professional misconduct lay with the Institute of Chartered Accountants of India, the statutory body created by the Chartered Accountants Act 1949, which both represented the profession and judged its members. The department administering the Companies Act could act on individual companies, ordering inspection or investigation of their affairs, but there was no public authority charged with inspecting audit quality, setting the terms of audit oversight or sanctioning firms as firms. The Naresh Chandra Committee recommended an independent oversight board in 2002; the recommendation was not implemented under the old Act. The Companies Act 2013 created the National Financial Reporting Authority to fill the gap, and it was constituted in 2018.