The Equal Remuneration Act 1976 is usually filed away as a well-intentioned statute that failed for want of enforcement, and that filing is a misdiagnosis. The Act was enforced about as well as its design allowed, because its design contained the failure inside it. Parliament required an employer to pay a woman no less than a man for performing the same work or work of a similar nature, and then defined that phrase so that the comparison only works where a man and a woman are already doing substantially the same job inside the same establishment. In a labour market where men and women are sorted into different occupations, different industries and different grades long before a pay packet is calculated, that comparator is exactly what a woman worker usually does not have. The statute therefore governed the rarest form of pay discrimination, the visible side-by-side kind, and left the commonest form, the segregation of women into lower-paid work, outside its reach altogether. Its successor, the Code on Wages 2019, widened the protected characteristic from sex to gender and rebuilt the machinery around the duty, and then kept the same test. Understanding this Act means understanding that drafting choice, because everything else about it, the advisory committees, the inspectors, the penalties, the thin line of cases, is commentary on a definition in Section 2(h).

The world the Equal Remuneration Act was passed into

The constitutional promise came first, and it came in the weaker of the two forms the Constitution offers. Article 39(d) directs the State to secure that there is equal pay for equal work for both men and women, and it sits in Part IV, among the Directive Principles of State Policy, which Article 37 declares fundamental in the governance of the country but not enforceable by any court. The Constituent Assembly had the option of placing equal pay among the justiciable fundamental rights and did not take it, so for the first quarter century of the Republic the principle operated as an instruction to legislatures rather than as a claim a worker could carry anywhere. What filled the space instead was administrative practice, and the practice was frequently discriminatory in terms. The Committee on the Status of Women in India, whose report Towards Equality was submitted at the end of 1974, documented the position in detail, including minimum wage notifications in which governments themselves fixed lower rates for women than for men for the same work, recording among its examples the Andhra Pradesh fixation of lower minimum wages for women in tobacco, zarda and cigarette work irrespective of the work the women did. A statute book that contained Article 39(d) coexisted with wage schedules, issued under other statutes, that wrote the differential into law.

The road from the Committee’s report to the Ordinance was short, and its shortness is part of the explanation of the text that emerged. Towards Equality was tabled in Parliament in February 1975, and after a protracted discussion of its findings the House adopted a resolution authorising the government to mount comprehensive legislative and administrative measures to remove the disabilities and discrimination the report had documented. The report had not confined itself to wages: it ranged across education, health, political representation and personal law, and its legislative harvest over the following years was correspondingly broad, reaching from this Act to the amendments of the dowry law in the following decade and the National Commission for Women at the start of the nineteen nineties. For remuneration specifically, the Committee’s finding carried a weight that a pressure group’s pamphlet would not, because the differentials it exposed were not only private arrangements but the product of official wage fixation, and a government confronted with its own committee’s demonstration that the State fixed lower minimum wages for women than for men for the same work was a government that had to legislate or explain itself in International Women’s Year. The Ordinance of 26 September 1975 was the immediate answer, and the Act of the following February its considered form; but the speed of the sequence, report to resolution to ordinance inside a single year, with a session of Parliament asked to replace an operating ordinance rather than to design a statute from first principles, belongs in any account of why the Convention’s standard, which would have required the design of an evaluation machinery no Indian institution then possessed, was not the standard the Bill proposed. There is a longer prehistory in the Constituent Assembly, where the equal pay clause had been debated a generation earlier and placed, to the recorded disappointment of members who had argued for it as a right, among the Directive Principles; the Assembly’s choice fixed the constitutional grammar within which everything later was written, a State directed toward equal pay, a worker armed with no direct constitutional claim to it, and a legislature whose eventual statute would be measured against a directive rather than against a right. The courts’ later deduction of an enforceable principle from Articles 14 and 16, examined below, was in substance an escape from that grammar, achieved by reading the directive back into the equality code; the statute of 1976 was written inside the grammar, as its fulfilment, and inherited its modesty.

The international obligation was older than the domestic one and wider than it. India ratified the Equal Remuneration Convention, 1951 (No. 100) of the International Labour Organization on 25 September 1958. Article 2 of that Convention commits a ratifying State to promote and, so far as consistent with national wage-fixing methods, to ensure the application of the principle of equal remuneration for men and women workers for work of equal value. The operative words are the last four. Equal value is a comparison between different jobs, a nursing aide against a storekeeper, a midwife against a fitter, assessed by the skill, effort, responsibility and conditions each job demands; same or similar work is a comparison between substantially identical jobs. India accepted the wider standard internationally in 1958 and legislated the narrower one domestically when it finally legislated, and the distance between those two formulations is the subject of this article. The field as a whole, of which this Act forms one part beside the maternity, harassment and criminal law statutes, is mapped in the guide to women’s rights legislation in India, and this article confines itself to the remuneration statute and what its text did.

Why did the Act begin as an ordinance in 1975?

The Equal Remuneration Ordinance, 1975 was promulgated by the President on 26 September 1975 because the government wished Article 39(d) to be implemented in the year being celebrated as International Women’s Year, and an ordinance was the instrument that let it act inside that calendar.

The Act that replaced the Ordinance in the following session carried the same explanation in its Statement of Objects and Reasons.

The route mattered beyond symbolism. Legislation by ordinance compresses the deliberative stages in which a definitional choice of this weight would ordinarily be argued over: there was no select committee stage at which the difference between the Convention’s equal value standard and the Bill’s same or similar work standard had to be defended against the comparison, and by the time Parliament considered the replacing Bill, the narrower formula was already the law in operation and the debate was about continuing it rather than choosing it. The Equal Remuneration Act, 1976 (Act 25 of 1976) received the assent of the President on 11 February 1976, and its Section 18 repealed the Ordinance, which was Ordinance 12 of 1975, while saving everything done under it. The Act’s long title states its two purposes plainly: to provide for the payment of equal remuneration to men and women workers, and for the prevention of discrimination, on the ground of sex, against women in the matter of employment. Both purposes were real. The question the rest of this article answers is why a statute with those purposes, an overriding-effect clause and a criminal sanction produced so little change in what women were paid.

The moment had a further feature that no account of the Act’s origins should omit. Both the Ordinance of September 1975 and the Act of February 1976 are instruments of the Emergency, the proclamation of June 1975 remaining in force until March 1977, and the period’s legislative style, government by ordinance followed by enactment in a Parliament whose deliberative life was under the shadow of the proclamation, is legible in the statute’s passage. The point is not that the Act was an Emergency measure in purpose; its subject was the Committee on the Status of Women’s report and International Women’s Year, and its contents were continuous with the labour legislation of ordinary times. The point is about scrutiny. A Bill whose central definitional choice departed from the international standard the Bill existed to implement was exactly the kind of measure on which a select committee, an alert opposition and a reporting press might have fastened in a normal session, and the conditions of 1975 and 1976 were not those of a normal session. The historian of the statute is entitled to record that the narrow test entered the law at the moment of the Republic’s weakest parliamentary scrutiny, and that it then acquired the immortality that first formulations in a field tend to acquire, being copied, as first formulations are, by the amenders of 1987 and the consolidators of 2019 who had other battles to fight. Path dependence is a complete explanation of much in this Act’s afterlife, and it begins here.

One further feature of the moment deserves recording, because it explains the Act’s second, quieter half. The statute was not only a pay statute. Its framers understood that a woman who cannot get hired into the better-paid grades has no use for an equal pay rule, and so the Act carried from the start a prohibition on discrimination in recruitment and a machinery of advisory committees charged with increasing employment opportunities for women. The design therefore recognised occupational segregation as a problem on its face, and then addressed it with the weakest instruments in the book, advice and exhortation, while reserving the enforceable duty for the comparison that segregation defeats. That pairing, a hard duty pointed at the symptom least often present and a soft instrument pointed at the cause most often present, is the Act in one sentence.

How the differential was administered before 1976

To see what the Act had to displace, it is necessary to see how women’s lower pay was produced in the decades before it, because it was rarely produced by an employer simply announcing two rates for one job. It was produced by the ordinary machinery of wage fixation operating exactly as designed. Under the Minimum Wages Act, 1948, governments fixed minimum rates for scheduled employments, and the schedules were commonly drawn by occupation and, within occupations, sometimes expressly by sex, so that the floor itself was lower for a woman than for a man at the same task; the Towards Equality report’s documentation of such fixations was damaging precisely because the discriminator in those instances was the State’s own notification. In organised industry, rates were set by collective bargaining and by the awards of industrial tribunals, and both inherited the job structure they found: occupations were classified, classifications carried rates, and the classifications into which women were concentrated, the lighter processes, the finishing and packing lines, the cleaning and auxiliary grades, carried the lower rates as a property of the classification rather than as a stated preference about women. Piece rates in agriculture and in home-based manufacture were fixed by custom and by the contractor’s margin. In each case the differential wore the clothing of some neutral instrument, a schedule, a classification, a custom, and in each case the consequence for the design of a remedy is the same. A statute aimed at the announcement, two rates for one job openly stated, would find almost nothing to strike at, because the announcement was rarely made. A statute aimed at the rate structure had to be able to look through the classification to the work, which is what the similar nature limb of Section 2(h) attempted, and to override the instruments that carried the rates, which is what Section 3 attempted. The framers understood the mechanism of the harm with some precision. Their failure was not a failure to see how the differential was administered; it was the narrower failure, examined below, to follow the analysis across the boundary between jobs, where the classification system did its deepest work.

The same period explains why the claim, when Parliament created it, was routed to the labour administration rather than to the ordinary courts or to a specialised tribunal of a new kind. Indian labour law’s settled pattern was the authority model: a named officer, drawn from the labour department, deciding claims summarily with civil court powers for evidence, an appeal inside the administration, and recovery through the revenue or the Section 33C machinery of the Industrial Disputes Act. The Minimum Wages Act and the Payment of Wages Act both worked this way, and the Equal Remuneration Act adopted the pattern entire, authorities, inspectors, registers and all. The pattern had real advantages for a wage claimant: no court fee culture, no pleading technicalities, a forum that sat in the district and knew the local establishments. It also had the pattern’s settled disadvantage, which the record of this Act exhibits in full: the authority is part of the same administrative world as the inspectorate that had tolerated the rates, its docket is dominated by minimum wage and payment claims of a more mechanical kind, and a claim that requires it to hold, in effect, that an employer’s whole classification of women’s work is a contravention is a claim of a different order from an arithmetical short payment. Machinery is never neutral about the claims it can comfortably decide, and the machinery chosen in 1976 was comfortable with everything about an equal pay claim except its essence.

How the Act is built

The Act is short, eighteen sections in three chapters, and its brevity is itself information: this is a single-duty statute, not a code. Chapter I contains the short title, the definitions and the overriding effect. Chapter II, headed Payment of remuneration at equal rates to men and women workers and other matters, contains the whole substantive law: the equal pay duty in Section 4, the recruitment prohibition in Section 5, the advisory committees in Section 6 and the claims authorities in Section 7. Chapter III is machinery: registers, inspectors, penalties, offences by companies, cognizance, rule-making, directions, the special cases carve-out, the declaration power and the repeal of the Ordinance.

Section 3 gives the Act overriding effect, providing that its provisions have effect notwithstanding anything inconsistent contained in any other law, or in the terms of any award, agreement or contract of service, whether made before or after the commencement of the Act, or in any instrument having effect under any law. The clause was not decoration. Wage rates in organised industry were set by settlements and awards negotiated between employers and unions, and a sex-differentiated scale embedded in a settlement was precisely the kind of instrument the Act had to defeat if it was to operate at all; within a decade the Supreme Court was required to say, in the Mackinnon Mackenzie litigation examined below, that it did. The Act also sits inside the wider architecture of Indian labour legislation, whose threshold and coverage questions, which statutes bind which establishments and which governments administer them, are treated in the Indian labour law framework guide; the coverage question matters here more than for most statutes, for a reason the commencement design creates.

How was the Act switched on, one employment at a time?

Section 1(3) provided that the Act would come into force on such date, not being later than three years from its passing, as the Central Government might by notification appoint, and that different dates might be appointed for different establishments or employments.

The Act was accordingly extended employment by employment across three years, beginning with employments in local authorities on 1 January 1976 under the Ordinance and reaching agriculture only on 26 January 1978 and community, social and personal services on 3 June 1978.

The definition of worker in Section 2(i) makes the consequence explicit: a worker is a worker in any establishment or employment in respect of which the Act has come into force. Coverage under this statute was therefore never a single fact about the Indian economy; it was a rolling fact about particular employments on particular dates, and for the first years of the Act’s life a woman’s right to equal remuneration depended on whether the notification series had reached her industry. The roll-out followed the organised sector first: hospitals, nursing homes and dispensaries were brought in on 27 January 1976, banks, insurance companies and other financial institutions on 8 March 1976, educational, teaching and training institutions and mines on 5 April 1976, and the textile employments in stages from July 1976. The employments in which women workers were most numerous and worst paid, agriculture above all, came near the end of the sequence, on 26 January 1978. A reader who asks why the Act’s early enforcement record is so thin should begin here rather than with the inspectorate: for much of its first three years the statute had not yet arrived in the workplaces where most women worked.

The appropriate Government under Section 2(a) is the Central Government for employments carried on by or under its authority, for railway administrations, banking companies, mines, oilfields, major ports and corporations established by a Central Act, and the State Government for every other employment, which divided administration of the Act along the standard line of Indian labour law and meant that the quality of enforcement varied with the labour department of the State in which the establishment sat. Section 2(d) defines man and woman as male and female human beings of any age, a definition with the incidental consequence that the Act’s protection was never confined to adult workers.

The equal pay duty in Section 4

Section 4(1) states the duty in comparative form. No employer shall pay to any worker, employed by him in an establishment or employment, remuneration, whether payable in cash or in kind, at rates less favourable than those at which remuneration is paid by him to the workers of the opposite sex in such establishment or employment for performing the same work or work of a similar nature. Four features of that sentence control everything that follows. First, the duty is comparative within an establishment or employment: the claimant measures her rate against the rate the same employer pays workers of the opposite sex in the same establishment, so a woman who is the only person, or the only woman, doing her work in her establishment has no statutory comparator, however badly she is paid relative to the market. Second, the duty attaches to remuneration, a word Section 2(g) defines widely, as the basic wage or salary and any additional emoluments whatsoever payable, either in cash or in kind, to a person employed in respect of employment or work done in such employment, if the terms of the contract of employment, express or implied, were fulfilled; allowances, bonuses and benefits in kind are inside the duty, not outside it. Third, the standard of comparison is rates less favourable, which directs attention to the whole package and its structure, not merely to the headline wage. Fourth, the duty is imposed in the negative form of a prohibition, enforceable through the claims authority and the criminal law, rather than as a freestanding right to a determined sum.

Why does the equal pay duty run in both directions?

Because Section 4 speaks of workers of the opposite sex, the duty protects a man paid less than a woman for the same work exactly as it protects a woman paid less than a man. The mischief ran overwhelmingly one way, but the text is symmetrical.

A male claimant before a Section 7 authority stood on the same words as a female one. The symmetry is worth pausing on because it reveals the drafting philosophy. Parliament did not enact a women’s welfare measure with a penal clause attached; it enacted a rule about rates, which is why the long title speaks of equal remuneration to men and women workers while the discrimination provisions that follow are addressed to discrimination against women specifically. The two halves of the Act are drafted in different registers, and the difference is deliberate. Pay differentiation is treated as a defect in the wage structure itself, whoever it disfavours in the individual case; exclusion from employment is treated as a wrong done to women as a class, because that is where the record, including the record the Towards Equality report assembled, located it.

Section 4(2) then bars the obvious evasion: no employer shall, for the purpose of complying with sub-section (1), reduce the rate of remuneration of any worker. Equalisation must be achieved by levelling up, never by levelling down, which converts the duty from a rule about relative treatment into a rule with a floor built into it. The provision matters practically as much as doctrinally. Without it, an employer confronted with a claim could comply at zero cost by cutting the men’s rate, leaving the claimant no better off and the workforce divided against the statute that produced the cut; with it, compliance costs money, which is the honest price of the discrimination the rates embodied, and the claimant’s victory is a real increase rather than a shared reduction.

Section 4(3) dealt with the transition. Where, in an establishment or employment, the rates payable before the commencement of the Act for men and women workers for the same work or work of a similar nature were different only on the ground of sex, the higher of two rates, or the highest of more than two, became the rate payable to both from commencement. The proviso then limited the sub-section’s generosity in time: nothing in it entitled a worker to revision of the rate for service rendered before commencement, so the Act equalised prospectively and did not reopen past pay packets. The drafting is careful and, within its own terms, complete. What it could not do, and what no amount of care inside Section 4 could have done, is manufacture a comparator where the establishment’s own division of labour had abolished one. That defect sits upstream, in the definition to which the whole of Section 4 is keyed.

Before leaving Section 4, the definition of remuneration itself repays a closer look, because it foreclosed the second most obvious evasion after levelling down. An employer who may not differentiate in the basic wage can, in a system of fragmented pay, differentiate everywhere else: in the dearness allowance computed as a percentage for one sex and a flat sum for the other, in the bonus, in the overtime roster, in the provision of housing, transport or meals in kind. Section 2(g) closes those doors in advance by defining remuneration as the basic wage or salary and any additional emoluments whatsoever payable, either in cash or in kind, and the words whatsoever and in kind do the closing. The definition’s closing words, if the terms of the contract of employment, express or implied, were fulfilled, locate the entitlement in the contract’s own economy: remuneration is what the worker would receive on fulfilling the contract, so that an employer cannot defeat the comparison by pointing to sums a worker failed to earn through absence the contract did not excuse, and equally cannot inflate the male comparator’s figure with payments that were not remuneration at all. The practical consequence for any audit under the Act was that the comparison had to be run over the whole emolument structure as the registers and wage sheets disclosed it, element by element, and the element-by-element structure of real pay packets is also why two workers on an identical basic rate could still stand in a Section 4 relationship of inequality, through the allowances one received and the other did not. The definition is, in this respect, the best-drafted provision in the Act: it anticipated the fragmentation of pay that Indian wage structures were already exhibiting in the nineteen seventies and that has only deepened since, and it is not accidental that the Code on Wages, when it came to restate the field, adopted the same strategy at greater length, defining wages by inclusion and exclusion across the whole package rather than by the basic rate alone.

Can a settlement or contract override the equal pay duty?

No. Section 3 gives the Act effect notwithstanding anything inconsistent in any other law or in the terms of any award, agreement or contract of service, and the Supreme Court applied that clause in Mackinnon Mackenzie to refuse an employer the benefit of a negotiated settlement whose fitment terms treated women stenographers less favourably than men doing the same work.

The point has a practical edge for any employer auditing a legacy pay structure. A differential does not become lawful because it was agreed, because a union signed it, because it predates the Act, or because it is embedded in a standing settlement that nobody has renegotiated since. The overriding clause reaches into the settlement and displaces the inconsistent term, and the employer’s obligation is to find the differential and remove it by raising the lower rate, not to wait for a claimant to find it first. Wage settlements in organised industry were the principal vehicle by which sex-differentiated rates were carried forward from the pre-Act period into the Act’s own lifetime, which is why the framers wrote Section 3 as widely as they did and why the courts, when the question reached them, gave it its full width.

The definition that decides cases

Section 2(h) defines same work or work of a similar nature as work in respect of which the skill, effort and responsibility required are the same, when performed under similar working conditions, by a man and a woman, and the differences, if any, between the skill, effort and responsibility required of a man and those required of a woman are not of practical importance in relation to the terms and conditions of employment. Every element of that definition is a litigation opportunity, and the history of the Act is the history of employers taking them. The comparison is between the work of a man and the work of a woman, so the statute presupposes both. The skill, effort and responsibility required must be the same, subject only to differences that are not of practical importance, a phrase that invites the employer to magnify any difference, a confidential file handled here, a night shift worked there, into a distinction of principle. The working conditions must be similar. And the whole inquiry is conducted job by job, so the statute has nothing to say about the relationship between the pay of two different jobs, however large the gap and however sex-segregated the two workforces performing them.

Set the three standards side by side and the drafting choice becomes measurable. Under a same work standard, the claimant must find a person of the opposite sex doing the identical job. Under a similar work standard, the job must be alike in skill, effort and responsibility under similar conditions, with residual differences dismissed only if they lack practical importance. Under an equal value standard, the claimant’s job need only be equal in value to a different job, assessed on the same factors of skill, effort, responsibility and conditions, so a comparison across an occupational boundary is possible in principle. Indian law adopted the first two and has never adopted the third. The Convention India ratified requires the third. The table below is the article’s findable artifact, because the entire fifty-year record of this statute is legible from it.

Standard	The test it applies	A comparison that succeeds	A comparison that fails	Who uses it
Same work	The two workers do the identical job, so the rates must match	A woman and a man working the same machine on the same shift in one establishment are paid different daily rates	The woman works a different machine of equal difficulty, so no identical job exists to compare	The first limb of Section 2(h) of the Equal Remuneration Act 1976
Work of a similar nature	Skill, effort and responsibility required are the same under similar working conditions, and any differences are not of practical importance	Women stenographers fitted into a common scale on worse terms than men stenographers, as in Mackinnon Mackenzie	The employer persuades the authority that a difference in duties, however small, is of practical importance to the terms of employment	The second limb of Section 2(h), and the test the Code on Wages 2019 retained
Work of equal value	Two different jobs are equal in value when rated on skill, effort, responsibility and working conditions, so their rates must match	A predominantly female job, such as nursing assistance, rated equal in value to a predominantly male job, such as storekeeping, yet paid less	No claim can fail for want of a comparator, though it can fail on the rating itself, which must be proved job by job	Article 2 of ILO Convention No. 100, ratified by India in 1958 and never enacted into domestic law

The contest over practical importance

The phrase on which claims under the Act were actually fought is the closing clause of Section 2(h): the differences between what the man’s work and the woman’s work require must not be of practical importance in relation to the terms and conditions of employment. The clause is a deliberate relaxation of identity, an acknowledgment that no two workers’ days are identical and that a test requiring identity would be unusable; but its effect in litigation was to hand the employer a standard-sized opening in every case. The employer’s argument wrote itself in three moves. First, enumerate the differences between the two jobs, and there are always differences: the man lifts the heavier crate on occasion, the woman handles the confidential file, the man’s line runs at night, the woman’s customers are of a kind said to need a woman’s manner. Second, characterise each difference as touching the terms and conditions of employment, since almost any difference can be connected, with ingenuity, to effort, responsibility or conditions. Third, invite the authority to hold that a difference so characterised is of practical importance, at which point the statutory comparison collapses and the rates may lawfully differ. The claimant’s answer also wrote itself, and it is the answer the Supreme Court accepted in Mackinnon Mackenzie: differences relied on must be real rather than constructed for the litigation, they must bear on the work rather than on the sex of the worker, and a difference that consists in the employer’s own practice of reserving a task or a title to one sex cannot be prayed in aid, because it is the discrimination describing itself. The confidential stenographer point in that case is the type specimen. The employer’s case depended on the women’s work being different in kind, confidential where the men’s was general; the Court’s response was that nothing in the nature of confidential stenography made it women’s work, and that if only women were doing it, that was the management’s choice, not a property of the job.

Two lessons about the statute’s real operation follow from this contest, and both are worth more than the doctrinal summary. The first is that Section 2(h) made every equal pay claim an evidence contest about job content, conducted before a Labour Officer-grade authority, in which the employer held most of the documents, the muster rolls, the duty charts, the classification records, and the claimant held her own testimony and whatever her union had gathered. A test that turns on fine discriminations of skill, effort and responsibility rewards the party with the better archive, and that party is rarely the claimant. The second is that the practical importance clause, for all its litigability, was not the Act’s deepest problem; it governed the margin of the similar work limb, the cases near the boundary of a comparison that exists. The deepest problem governed the cases where there was no comparison to have a margin on. Commentators who blame the clause for the Act’s record mistake the skirmish for the war. Even a claimant who won every practical importance argument in the country would have won only inside establishments where men and women already did the same work, and the pay gap did not live principally in those establishments.

What the difference means for a claimant can be stated as a rule of thumb. The narrower the standard, the more the case turns on the architecture of the employer’s own workforce, a matter the employer controls; the wider the standard, the more the case turns on the content of the jobs, a matter that can be proved by evidence and contested by expertise. An employer who wishes to be safe under a same or similar work regime need only organise production so that men and women rarely do the same work, which is, in most industries, how production was already organised. An employer under an equal value regime gains no such shelter from segregation, because segregation is the very thing the comparison is designed to see through. This is why the definitional question is not a technicality and why reform proposals in this field are always, at bottom, proposals about Section 2(h) and its successors.

For completeness, the elements of a claim under the Act can be stated as a sequence, because a claimant who could not establish every link in it had no case however unequal her pay. She had to show an establishment or employment to which the Act had been extended by notification on the relevant date; a worker of the opposite sex employed by the same employer in that establishment or employment; the same work or work of a similar nature performed by both, within the meaning of Section 2(h); remuneration paid to her at rates less favourable than the comparator’s, computed over the whole emolument structure Section 2(g) defines; and a difference not explained by a factor other than sex, not saved by the special cases in Section 15, and not certified away by a declaration under Section 16. Each element was a point of failure, and they failed in a characteristic order: the claim usually died at the second or third element, on the comparator or on similarity, long before it reached the arithmetic of the fourth. A statement of the elements in this form is also the cleanest way to see what the Code changed and what it did not, for the Code’s Section 3 claim has the same skeleton, the same comparator, the same similarity inquiry under its Section 4, with the elements of remedy and time rebuilt around it. Statutes in this field do not usually fail at their prohibitions. They fail at their elements, and the elements are where a reader should always look first.

The recruitment prohibition and its boundaries

Section 5, as enacted, provided that on and from the commencement of the Act, no employer shall, while making recruitment for the same work or work of a similar nature, make any discrimination against women, except where the employment of women in such work is prohibited or restricted by or under any law for the time being in force. Two features of the original text deserve emphasis. The prohibition was addressed to the gate, not to the corridor: it governed the moment of hiring and said nothing, in terms, about what happened to a woman after she was hired, so an employer who recruited women without discrimination and then channelled them into the lowest grades, denied them the training that led to promotion, or transferred them out of the better-paid lines was outside the section’s literal reach. And the prohibition was keyed to the same comparison as the pay duty, recruitment for the same work or work of a similar nature, so it protected a woman’s access to jobs that men also did, and had less to say about jobs from which women were excluded altogether and which therefore had no male incumbents at all. The proviso protected priorities and reservations in recruitment for the Scheduled Castes and Scheduled Tribes, ex-servicemen, retrenched employees and other classes or categories of persons, a necessary saving which prevented the sex discrimination ban from being read as a solvent of the reservation system.

The exception for work from which women’s employment is prohibited or restricted by law was not hypothetical when the Act was passed. Indian labour law carried a stock of protective restrictions on women’s work, limits on night work and on underground and hazardous employment descended from an older protective philosophy, and each such restriction operated under Section 5 as a lawful reason to refuse a woman the job and its rate. The interplay mattered in both directions. Protective law kept women out of some of the better-paid shifts and processes, which depressed their earnings by exclusion; the equal remuneration statute then declined to treat that exclusion as discrimination, because the excluding law was its own exception. A complete account of the period’s law has to hold both halves together: the same legal system that enacted equal pay maintained, elsewhere in the book, a set of sex-based exclusions from employment that the equal pay statute expressly declined to disturb.

The protective stock the exception preserved is worth enumerating, because its earnings consequences were large and largely invisible in equal pay debates, which fix on rates. Section 66 of the Factories Act, 1948 confined the employment of women in factories to the hours between six in the morning and seven in the evening, subject to limited State variation and an absolute bar on night work in the statutory scheme as it stood for most of the Act’s life, which excluded women from the shifts and the continuous-process rosters to which shift allowances and overtime earnings attached. The Mines Act, 1952 prohibited the employment of women underground altogether and restricted their employment above ground to the daylight hours, removing them from the better-paid underground grades of the industry the notifications had reached in April 1976. State shops and establishments legislation and industry-specific rules added their own limits in places. Each restriction was defended, in its own terms, as protection, of health, of safety, of morals as the period understood them, and some part of the defence was genuine; but the aggregate effect was a statutory ceiling on women’s access to the hours and processes in which earnings are made, operating entirely outside the remuneration statute, which by Section 5’s exception expressly declined to treat exclusion under such laws as discrimination. A woman’s pay packet could therefore be smaller than a man’s, lawfully and substantially, without any rate in any register differing, because the composition of her working week had been fixed by a different statute book. This is the second sense in which the Act’s field of operation was narrower than its title promised: it governed the rate for the work, and a body of protective law governed access to the work, and the boundary between the two was the boundary of its exception clause.

Two later provisions mark the boundaries of the whole scheme. Section 15, as substituted by the amending Act of 1987, provides that nothing in the Act applies to cases affecting the terms and conditions of a woman’s employment in complying with the requirements of any law giving special treatment to women, or to any special treatment accorded to women in connection with the birth or expected birth of a child, or with terms and conditions relating to retirement, marriage or death. The carve-out is the statute’s own acknowledgment that formal equality of rates coexists with a body of law that treats women differently for protective and maternity purposes, and that the Act was not to be used as an argument against the second body of law by, for example, an employer resisting a maternity obligation on the ground that men received no equivalent. Section 16 is the more consequential boundary. Where the appropriate Government is, on a consideration of all the circumstances of the case, satisfied that the differences in regard to the remuneration, or a particular species of remuneration, of men and women workers in any establishment or employment are based on a factor other than sex, it may, by notification, make a declaration to that effect, and any act of the employer attributable to such a difference shall not be deemed to be a contravention of any provision of the Act. The declaration power is a safety valve of unusual width: it lets the executive certify an entire differential out of the statute, establishment by establishment, on the ground that some factor other than sex explains it. Factors other than sex, seniority, merit, the quantity or quality of output, the incidents of a genuinely different job, are of course legitimate explanations of pay differences, and every equal pay regime needs a way of recognising them; the ordinary way is to let the employer prove the factor in the individual case before the deciding authority. Section 16 instead made the explanation a matter for governmental declaration in advance, a device that converts a question of evidence into a question of administrative satisfaction, and its existence in the Act is a standing reminder that the statute’s framers built an exit for the difficult cases as carefully as they built the duty for the easy ones.

The declaration power deserves a further word, because it interacts with the claims machinery in a way a reader of Section 7 alone would miss. A claim before the authority turns on evidence: the rates, the work, the factors said to explain the difference. A declaration under Section 16 short-circuits that inquiry at the level of the establishment, for if the appropriate Government has declared a difference in an establishment or employment to be based on a factor other than sex, the employer’s acts attributable to the difference are deemed not to be a contravention, and there is nothing left for the authority to try. The power is not in terms confined to differences already examined in a claim, nor is it expressed to require a hearing of the workers affected, and it is exercisable on the Government’s satisfaction on a consideration of all the circumstances, a formula that commits the merits to the executive. No reported body of Section 16 declarations works its way through the case law of the Act, and the power’s significance is therefore constitutional rather than operational: it marks the point at which the Act’s own scheme allowed the political branch to relieve an employer of the duty wholesale, in advance, on an explanation the workers had no statutory right to test. Set beside Section 3, which displaces even settlements in the duty’s favour, Section 16 completes the statute’s internal tension. The Act overrides private arrangements to protect the duty, and reserves a public power to dissolve the duty where the Government is satisfied the difference is innocent. Both provisions were enacted in the same text, by the same Parliament, and the distance between them is a fair measure of the distance between the Act’s announced purpose and its operative temper.

The duty at work: three workplaces

The operation of the test is best seen in application, and three workplaces, composites drawn from the patterns the case law and the reports record, show the statute succeeding, straining and failing on identical facts of inequality.

The first is a textile mill in which men and women work the same looms in the same shed, the men on a daily rate fixed by the applicable award and the women, engaged through a labour contractor, on a lower daily rate for the same output. This is the Act’s home ground. The work is the same work within Section 2(h) without argument; the establishment is one; the only defence available is that the women are not the mill’s workers but the contractor’s, which goes to the identity of the employer rather than to the comparison, and the claim before the Section 7 authority is otherwise an arithmetical exercise on the muster roll and the contractor’s bills. Claims of this pattern were the claims the Act could decide, and differentials of this pattern, in the organised employments the notifications reached first, were the differentials the Act genuinely suppressed. It is worth noticing what makes the case easy: a male incumbent doing the identical job, in the same establishment, at a rate recorded in a document the inspector can seize. Every element of that sentence is a condition, and the Act’s history is what happened when one or more of them was absent.

The second workplace is a pharmaceutical plant in which the production lines are staffed by men and the packing hall exclusively by women, the packing work requiring sustained dexterity, attention to labelling accuracy on which regulatory liability turns, and a full shift on foot, and paid at a little over half the production rate. The women have no claim under Section 4. There is no man performing the same work or work of a similar nature in the establishment; the packing job’s skill, effort and responsibility are not the same as the production job’s, because they are different in kind, and the comparison the statute offers simply does not engage. That the packing work is undervalued relative to its demands, that it is staffed by women because it is undervalued and undervalued because it is staffed by women, and that a job evaluation rating the two jobs on common factors might well place them within a grade of each other, are propositions the Act has no machinery to receive. This is the equal value case, and it is not an exotic construction: it describes the ordinary structure of a great part of manufacturing employment, in which the sex segregation of processes is the plant’s organising fact. The women’s remedies lie elsewhere, in minimum wage fixation for the packing employment, in union bargaining over the classification, in the slow politics of regrading, and none of them is a right against the differential as such.

The third workplace is a bank in which a woman officer is paid on the same scale as her male colleagues, as the industry’s settlements have long provided, but has been passed over for the postings that carry the allowances and the promotion prospects, the branch managerships and the inspection tours, which go to men on stated grounds of mobility and availability. Here the pay duty is satisfied and the grievance is real, and the Act’s answer lies in Section 5 as amended in 1987: discrimination in a condition of service subsequent to recruitment, in promotion and transfer, is prohibited, and the woman’s complaint to the authority is about the pattern of postings rather than about a rate. The case shows the amended Act working on its second limb, and also shows that limb’s limits: the authority can direct that adequate steps be taken to end the contravention, a remedial formula far less exact than the payment of a computed difference, and the individual posting decision, defended on the facts of the individual case, is harder to displace than a rate in a scale. Between the three workplaces the whole statute is visible: exact where a rate and a comparator coexist, silent where segregation has removed the comparator, and general where the wrong is a pattern of decisions rather than a number in a register.

A fourth workplace tests a boundary the other three do not reach. A woman is appointed to a post, a section officer in a public undertaking, a senior clerk in a private firm, at the starting rate of the scale; her predecessor in the identical chair, a man, had by the end of his tenure drawn the top of the scale, and, the sharper variant, a man appointed to the same chair at the same time in a sister unit is started two increments higher on a stated assessment of his experience. The European jurisprudence the Supreme Court surveyed in Mackinnon Mackenzie had already confronted the predecessor variant, in a case concerning a woman paid less than the man who had held her job before her, and had held that the equal pay principle in the European treaty was not restricted to contemporaneous employment, since otherwise an employer could defeat it by the simple expedient of never employing the two sexes in the post at the same time. The Indian text is less accommodating on its face: Section 4 compares the claimant’s rates with those at which remuneration is paid to workers of the opposite sex in the establishment, a present-tense comparison with the establishment’s workforce, and a predecessor who has left is a worker of the establishment no longer. Whether an Indian authority would read the section to admit the predecessor as comparator, treating the comparison as one between the rate for the work and the rate for the work, or would confine it to the serving workforce, is a question the reported case law under the Act does not answer, and it is stated here as open rather than resolved. The contemporaneous variant of the fourth workplace is not open at all: two appointees of opposite sexes, started at different points of one scale for the same work on a justification of experience that does not survive examination, present the cleanest Section 4 case there is, and the employer’s experience justification, to hold, must be a factor other than sex actually reflected in the two records. The distance between the two variants is a small map of the statute’s temperament: exact about numbers in a register, hesitant about time, and dependent throughout on the accident of who else the establishment happens to employ.

A fifth workplace is the field, and it is where most women workers in India were to be found throughout the Act’s life. Agricultural operations are traditionally sex-typed in their assignment: ploughing and irrigation are men’s work in the received division, transplanting, weeding and harvesting are shared or women’s, and the daily wage attaches to the operation rather than to the worker, so that the operations women perform carry the lower rate as a property of the task list. Where the rate for transplanting is lower than the rate for ploughing, the woman transplanting beside the man ploughing is not, in the statute’s terms, doing the same work or work of a similar nature, for the skill, effort and responsibility of the operations differ in kind, and the comparison fails at the threshold although the day’s labour is equally long and the employer’s establishment, a landholder’s fields worked by a gang engaged by the season, is the same ground. Two features compound the failure. The wage in much agricultural employment was, and in part remains, fixed or shadowed by the State’s own minimum wage notifications, which for years were drawn by operation and in places by sex, so that the differential the Act forbade an employer to practise could simultaneously be the differential the schedule prescribed, and an authority hearing a claim was in substance being asked to hold the notification’s structure unlawful under a statute that proceeded by a different department’s hand. And the Act’s arrival in agriculture came only on 26 January 1978, late in the notification sequence, into the employment where its institutional presuppositions, an identifiable employer, a register, a muster roll, an inspector’s visit, fitted least well of all. The field is therefore the limiting case of the whole design, the workplace where the comparator test, the establishment model and the enforcement machinery each reach their weakest together, and any verdict on the Act that is drawn from the organised sector’s record alone should be discounted by the size of the workforce the field represents.

The machinery: committees, authorities, inspectors

Section 6 obliged the appropriate Government to constitute one or more advisory committees for the purpose of providing increasing employment opportunities for women, to advise it with regard to the extent to which women may be employed in such establishments or employments as may be specified. Every committee was to consist of not less than ten persons nominated by the appropriate Government, of whom one-half were to be women. In tendering its advice, a committee was to have regard to the number of women employed in the concerned establishment or employment, the nature of the work, hours of work, the suitability of women for employment, the need for providing increasing employment opportunities for women including part-time employment, and other relevant factors. The appropriate Government, after considering the advice and after giving the persons concerned an opportunity to make representations, could issue such directions in respect of the employment of women workers as it thought fit. The design is characteristic of its period: a deliberative body, representative by nomination, charged with a promotional purpose, armed at the end of the process with a governmental power of direction whose breach Section 10(2) made an offence. What the committees were not given was any connection to the pay duty. They advised on the extent to which women might be employed, not on what women were paid, and so the one institution the Act created to look at women’s position in an industry was directed to the question of numbers rather than the question of rates. The advisory committee is the Act’s segregation instrument, and it is advisory in the strict sense: its product was advice to the government, and the government’s product, if any, was a direction whose enforcement depended on the same inspectorate as everything else. The committee idea was carried into subordinate legislation in its own right, the Central Advisory Committee on Equal Remuneration Rules of 1991 regulating the committee at the Centre, its composition, tenure and procedure, and State Governments constituted their counterparts within their spheres. As an institution the advisory committee belongs to a recognisable family in Indian labour governance, the tripartite and advisory body through which the State consults the interests before it directs, and its weakness in this statute was not the family weakness of talk without end. It was the specificity of its charge. A committee asked to advise on the extent to which women may be employed is a committee asked, in the idiom of its time, how far a protective boundary should move; it is not asked, and the Act nowhere asks it, whether the women already employed are paid what the men are paid, although that was the question the rest of the Act existed to answer. No reported line of Section 6(5) directions reshaping an industry’s employment of women stands in the Act’s record, and the committee system is best understood as the statute’s gesture toward the segregation problem, sincerely meant and structurally incapable of reaching it, since advice about the numbers of women to be employed in an establishment is not a lever on the valuation of the work those women do.

Section 7 created the forum. The appropriate Government could, by notification, appoint such officers, not below the rank of a Labour Officer, as it thought fit, to be the authorities for hearing and deciding complaints with regard to the contravention of any provision of the Act, and claims arising out of non-payment of wages at equal rates to men and women workers for the same work or work of a similar nature, defining the local limits of each authority’s jurisdiction. In the central sphere the appointed authorities were the Assistant Labour Commissioners, and inspectors under Section 9 were the Labour Enforcement Officers, so the Act was administered by the general labour administration rather than by any specialist body of its own. Section 7(3) concentrated the decisive question in the authority’s hands: if any question arises as to whether two or more works are of the same nature or of a similar nature, it shall be decided by the authority. After hearing the applicant and the employer and holding such inquiry as it considered necessary, the authority could direct, in a claim, that payment be made to the worker of the amount by which the wages payable exceeded the amount actually paid, and in a complaint, that adequate steps be taken by the employer to ensure no contravention. Every authority had the powers of a civil court under the Code of Civil Procedure, 1908 for taking evidence, enforcing the attendance of witnesses and compelling the production of documents, and was deemed a civil court for the purposes of Section 195 and Chapter XXVI of the Code of Criminal Procedure, 1973. An appeal lay, within thirty days of the order, to an appellate authority specified by notification, which could confirm, modify or reverse the order, with no further appeal; the appellate authority could allow a late appeal within a further thirty days where the appellant had been prevented by sufficient cause, but not thereafter. Amounts due under an authority’s decision were recoverable through the machinery of sub-section (1) of Section 33C of the Industrial Disputes Act, 1947, which Section 7(8) expressly applied.

Two silences in Section 7 are as significant as its contents. The section awards the claimant the difference between what was paid and what should have been paid, and nothing more: there is no compensation multiplier, no interest provision in terms, no damages for the discrimination itself, so the employer’s worst outcome on a claim was to pay, late, exactly what the Act had required it to pay on time. And the Act prescribes no limitation period for making a claim or a complaint, a point on which summaries of the Act are often wrong; the only periods the Act fixes are the thirty days for an appeal and the further thirty days for a delayed one. A general account of this kind can describe the route a claim took, but the route’s detail, the prescribed manner of application, the local limits of the authority for a given establishment and the appellate authority notified for it, had to be checked in the rules and notifications governing the individual case, and anyone reconstructing a historical claim or advising on a live matter under the successor Code should treat the general position stated here as a map rather than as advice on that case.

Around the authorities stood the preventive machinery. Section 8 required every employer, on and from the commencement of the Act, to maintain such registers and other documents in relation to the workers employed by him as may be prescribed, the register prescribed under the Equal Remuneration Rules being maintained in Form D, so that the rates paid to men and women were a matter of record and inspectable rather than of recollection. Section 9 empowered the appropriate Government to appoint inspectors, who were deemed public servants within the meaning of Section 21 of the Indian Penal Code, and who could enter any building, factory, premises or vessel at any reasonable time, require the production of registers, muster rolls and other documents and examine them, take evidence on the spot, examine the employer, his agent or servant and any person believed to be or to have been a worker, and make copies or take extracts. Section 11 extended liability for offences to companies, deeming every person in charge of and responsible for the conduct of the business guilty alongside the company, subject to the defence that the offence was committed without his knowledge or that he had exercised all due diligence, and reaching directors, managers, secretaries and other officers where the offence was committed with their consent or connivance or was attributable to their neglect.

Section 10 graded the criminal consequences, in the form the penalties took after the 1987 amendment discussed in the next section. Failures of record and cooperation, omitting to maintain a register, omitting or failing to produce a register or muster roll, omitting or refusing to give evidence or information, or preventing another from giving evidence, were punishable with simple imprisonment extending to one month, or a fine extending to ten thousand rupees, or both. The substantive offences were treated more seriously: making a recruitment in contravention of the Act, making a payment of remuneration at unequal rates to men and women workers for the same work or work of a similar nature, making any discrimination between men and women workers in contravention of the Act, or failing to carry out a direction under Section 6(5), attracted a fine of not less than ten thousand rupees and extending to twenty thousand rupees, or imprisonment for a term of not less than three months and extending to one year, or both, for the first offence, and imprisonment extending to two years for the second and subsequent offences. A person who, being required to do so, refused to produce a register or document to an inspector or to give information was liable to a fine extending to five hundred rupees, a figure that survived unrevised in Section 10(3) and gives the scale of the original scheme more honestly than the amended figures do. Section 12, in its substituted form, confined cognizance to courts not inferior to a Metropolitan Magistrate or a Judicial Magistrate of the first class, and to complaints made by the appropriate Government or an authorised officer, or by the person aggrieved or a recognised welfare institution or organisation.

The Act in the workplace: the claim in practice and the employer’s duties

A claim under the Act began, in the manner the rules prescribed, with an application to the authority appointed for the local limits within which the establishment sat, and its anatomy differed from a suit in every way that mattered to a worker deciding whether to file it. There was no plaint to draft and no valuation to defend; the application stated the establishment, the work, the comparator and the rates, and the authority’s inquiry, conducted with the powers of a civil court for evidence and attendance, did the rest. The decisive evidence was documentary and it was the employer’s: the registers Section 8 required, maintained in the prescribed form, the muster rolls, the wage sheets, the settlement or award that fixed the scales. The inspectorate’s parallel power to enter, inspect and copy meant that the material on which the comparison turned could be obtained by the State’s own officer without the claimant litigating for discovery. In design, therefore, the Act answered the information problem better than most labour statutes of its generation: the records duty, the inspection power and the summary forum were a coherent set. In operation the coherence was dissipated by the ordinary conditions of the labour administration, districts in which a single authority carried the claims of several statutes, inspections that were announced by the rhythm of the office rather than by the risk of the establishment, and registers maintained in the form that satisfied the visiting inspector rather than in a form that made a differential legible. The lesson the Act’s record teaches about machinery is the unfashionable one that design coherence is a necessary condition of enforcement and nowhere near a sufficient one.

The criminal side of the Act illustrates the same gap between text and operation. Until the amendment of 1987, a prosecution could be set in motion only on the complaint of the appropriate Government or an officer it authorised, which placed the initiative in the hands of the department whose inspectorate had already passed the establishment, and the fines on conviction were the one thousand and five thousand rupee figures of the original Section 10. A prosecution was therefore a departmental confession as much as an enforcement act, and it is not surprising that the Act’s criminal record is so much thinner even than its claims record. The 1987 substitution of Section 12, admitting the complaint of the person aggrieved and of a recognised welfare institution or organisation, was Parliament’s acknowledgment of the blockage, and the recognition of organisations working among women workers, the Centre for Women’s Development Studies, the Self-Employed Women’s Association, the Working Women’s Forum and the Institute of Social Studies Trust among the earliest recognised in the central sphere, created a channel by which a claim could be carried by an institution with the stamina for it. That the channel exists and was used sparingly says less about the organisations than about the arithmetic set out earlier: a channel is only worth entering when what flows through it is worth the journey, and the Act’s remedies, restitution for the claimant and a modest fine for the State, priced the journey high and the destination low.

For the employer, compliance with the Act was, in strictness, a programme with five elements, and setting them out exposes how far ordinary practice stood from it. The first was an audit of rates by job rather than by cadre, since the Act’s question is whether the work is the same or similar, and a pay structure organised by cadre titles, men’s grades and women’s grades, could conceal a contravention inside its own vocabulary until someone translated the titles back into duties. The second was the fitment discipline that Mackinnon Mackenzie imposed: where scales were common, the terms on which workers entered and moved within them had to be sex-neutral in operation and not merely in print. The third was the maintenance of the prescribed registers in a state that would bear inspection, because the register was both a duty and the employer’s own best evidence that its differentials rested on factors other than sex. The fourth, after 1987, was the extension of the audit from rates to decisions: recruitment shortlists, training nominations, promotion boards and transfer patterns all fell within Section 5, so that a compliance review confined to the wage bill was reviewing the smaller half of the exposure. The fifth was attention to settlements, since Section 3 displaced inconsistent terms in awards and agreements however venerable, and an employer who renewed a settlement without auditing its sex-differentiated survivals renewed the contravention with them. It is a measure of the Act’s practical standing that this programme, none of it esoteric, describes the practice of a minority of establishments across the Act’s lifetime; the statute asked to be complied with in advance, and was mainly encountered, by employers and workers alike, in retrospect.

The amendment of 1987: wider words, heavier penalties

The Act was amended once, substantially, by the Equal Remuneration (Amendment) Act, 1987 (Act 49 of 1987), which came into force on 16 December 1987, and the amendment is the point at which Parliament had the opportunity to reconsider the definition and chose instead to repair everything around it. The amendment’s first change was to Section 5: it inserted into the recruitment prohibition the words covering any condition of service subsequent to recruitment, such as promotions, training or transfer, so that the discrimination ban followed a woman through her employment instead of stopping at the door. The change was an implicit admission that the 1976 text had mislocated much of the discrimination it existed to prevent; a woman’s pay over a career is determined less by the rate on the day she is hired than by the grades she is allowed to enter, the training she is given and the lines she is transferred into, and the original Section 5 had left each of those decisions outside itself.

The second change was to the penalties. As enacted in 1976, the fine for the record offences had extended to one thousand rupees and the fine for the substantive offences to five thousand rupees, figures which, even at the price level of the nineteen seventies, priced a contravention below the saving a differentiated wage bill produced in a month. The amendment substituted the scale set out above: simple imprisonment extending to one month or a fine extending to ten thousand rupees for the record offences, and, for unequal payment, discriminatory recruitment and the other substantive offences, a mandatory minimum fine of ten thousand rupees rising to twenty thousand, or imprisonment of not less than three months extending to one year, or both, with imprisonment extending to two years for a second or subsequent offence. A mandatory minimum sentence of imprisonment in a labour statute is a serious legislative signal, and its presence in the amended Section 10 shows that by 1987 Parliament understood the original scheme to have been treated as a licensing fee rather than a prohibition.

The third change opened the criminal door wider. Section 12 was substituted so that cognizance could be taken not only on the complaint of the appropriate Government or its authorised officer, the gatekeeping arrangement of the original Act, but also on the complaint of the person aggrieved by the offence or of a recognised welfare institution or organisation, recognised for the purpose by the Central or a State Government; organisations such as the Centre for Women’s Development Studies in New Delhi, the Self-Employed Women’s Association in Ahmedabad and the Working Women’s Forum in Madras were among those recognised under the section. The amendment also substituted Section 15 in the form described above. What the amendment did not touch is the point. Section 2(h) survived 1987 exactly as it had been drafted in 1975, and so did the comparative structure of Section 4. Parliament widened the discrimination ban across the whole of a woman’s service, multiplied the fines, introduced minimum imprisonment and admitted the aggrieved worker and her organisations to the criminal process, and left the test that decided whether a pay claim could begin untouched. The amendment trail of this statute is therefore a study in the politics of the visible defect: recruitment discrimination and derisory penalties could be named and repaired in a way that commanded agreement, while the standard of comparison, whose repair would have required the State to commit itself to valuing women’s work against men’s across occupational lines, was left as it was.

The courts: a constitutional doctrine and a thin statutory docket

The most important fact about equal pay litigation in India is that almost none of it is about this Act, and most of it is not about women. The doctrine of equal pay for equal work was built by the Supreme Court as a branch of constitutional law, out of Articles 14 and 16 read with Article 39(d), in litigation brought overwhelmingly by public employees seeking parity with other public employees: regular against temporary, one department against another, one cadre against its neighbour. Part of the explanation is jurisdictional, and it is worth stating before the cases themselves. A public employee denied parity had no need of the Act at all: Articles 14 and 16 could be carried directly to a High Court or the Supreme Court by writ petition, a forum faster, more visible and more generous in remedy than a Labour Officer’s inquiry, and one in which the State as employer could be held to its own constitutional standard without any statutory comparator being located first. The constitutional route thus drew off exactly the class of claimant, the organised public employee with a union and a legal fund, who might otherwise have populated the statutory forum and generated its case law, and it left the Act to the private-sector and unorganised claimants least equipped to use it. Doctrines grow where their litigants are, and equal pay doctrine grew in the writ courts, about public employment, for this institutional reason as much as for any reason of legal logic. The point also explains a persistent confusion in the secondary literature, where the parity cases are cited as though they were decisions under or about the Equal Remuneration Act. They are not, and the distinction is not pedantry: a reader who believes Randhir Singh construes the Act will look in it for the Act’s test and find instead a constitutional standard that floats free of Section 2(h), available against the State and unavailable, in that form, against the private employer who was the Act’s principal subject.

In Randhir Singh v. Union of India, (1982) 1 SCC 618, the petitioner was a driver in the Delhi Police who claimed the pay of drivers in other government departments, and the Court held that the principle of equal pay for equal work is not an abstract doctrine but a constitutional goal deducible from Articles 14 and 16 construed in the light of the Preamble and Article 39(d), applicable where those drawing different scales of pay do identical work under the same employer and the differentiation rests on no classification or on an irrational one. The line continued through the nineteen eighties in cases about the engagement status of the claimant rather than the sex of the claimant: Dhirendra Chamoli v. State of Uttar Pradesh, decided in 1986, concerned casual employees of the Nehru Yuvak Kendras claiming the wages of regular employees doing the same work, and Surinder Singh v. Engineer-in-Chief, CPWD, decided the same year, held daily-wage workers entitled to the same wages and allowances as regular employees performing the same work. The doctrine’s modern restatement, in State of Punjab v. Jagjit Singh, (2017) 1 SCC 148, held that temporary employees, daily wagers, casual and contractual workers, performing the same duties and responsibilities as regular employees are entitled to wages at the minimum of the regular pay scale, and organised the whole line of authority around the question whether the duties are the same or similar in nature. The phrase will be recognised. The constitutional courts, working without a statute, converged on the same comparator-based test that Parliament had written into Section 2(h), and applied it to the comparison that public employment makes easy, between two groups of government servants doing the same job on different terms of engagement.

Against that large constitutional edifice, the Act’s own reported case law is remarkably small, and its leading case can be told in full. Mackinnon Mackenzie and Co. Ltd. v. Audrey D’Costa, (1987) 2 SCC 469, decided on 26 March 1987, arose under the Act itself. The employer, a shipping company in Bombay to which the Act had been applied with effect from 8 October 1976, employed stenographers, the men classified among the clerical and subordinate staff and the women engaged as confidential lady stenographers. A settlement placed both in a common scale of pay but fitted the women into it on terms less favourable than the men’s, so that women stenographers with the same length of service received less than men stenographers in the same scale. The claim travelled through the authority and the appellate authority under the Act and the High Court before reaching the Supreme Court, which dismissed the employer’s appeal. The Court held that the discrimination had been brought about by the terms of the settlement only on account of the sex of the employees and could not persist in view of Section 4; that the work of confidential lady stenographers was not a sex-based occupation, there being no custom or rule that only women could be confidential stenographers, and that if only women were doing the work it was because the management wanted them there; that a practice of appointing women as confidential stenographers could not be relied on to deny them the equal remuneration the Act required; that the management was liable to pay the same remuneration to all the stenographers on the same basis irrespective of sex, computed in accordance with the terms applicable to the male stenographers; and that the employer could derive no benefit from Section 4(3) and its proviso, because that sub-section addresses the case of different scales of pay before commencement, whereas here there had been one common scale and the inequality had been manufactured inside it at the stage of fitment. The judgment also surveyed the Equal Remuneration Convention and the English Equal Pay Act 1970 at some length, and the survey is the most interesting part of it for the argument of this article: the Court saw the wider world of equal remuneration law, including the European practice of comparing a woman’s pay with that of a man who had held her job before her, and then decided the case, correctly, on the narrower statute Parliament had given it.

The case repays attention for what it shows about the Act working as designed. The claim succeeded because every element the statute required happened to be present: one establishment, one employer, a single job performed by both sexes, a common scale, and a written instrument, the settlement, in which the differentiation could be read. Remove any one element, divide the women into a separate job title with no male incumbents, or set the rates in different establishments of the same group, and the same inequality becomes unreachable by the same section. Audrey D’Costa’s case is celebrated in the textbooks as the Act’s great victory, and so it was; it is also, in the half century of the Act’s operation, very nearly the whole of its Supreme Court record, which is the measure of the victory’s limits. The one other early appearance of the Act at that level is of a different character: in People’s Union for Democratic Rights v. Union of India, decided in 1982 and arising from the construction works for the Asian Games, the Supreme Court recorded, among a catalogue of labour law violations by contractors on public works, that women workers were being paid less than men for the same work contrary to the Equal Remuneration Act, treating the breach as one item in an indictment of enforcement rather than as a claim to be decided. Between a fitment dispute in a shipping office and an observation in a public interest petition, the statutory jurisprudence of equal pay in India was largely exhausted, while the constitutional jurisprudence of parity grew to fill volumes. A researcher who sets out to learn Indian equal pay law from the cases will learn a doctrine about government servants and engagement status; the statute about men and women will be a footnote in their notes, which is a fact about the statute and not about the researcher.

The constitutional doctrine also has a qualifying side, and an honest account of the courts must include it, because the qualifications are where the doctrine’s limits, and its resemblance to the statute, become exact. The Supreme Court never held that equal pay follows from similarity asserted at a high level of generality. In Grih Kalyan Kendra Workers’ Union v. Union of India, decided in 1991, the workers of a welfare organisation’s centres claimed parity with government employees, and the Court declined to decree equality on the material before it, holding that the question whether two sets of employees stand on the same footing in duties, responsibilities and qualifications is a factual inquiry unsuited to assumption, and remitting the matter for expert examination. The same insistence runs through the later cases: differentials grounded in genuinely different qualifications, functions or responsibilities survive challenge, because Articles 14 and 16 condemn irrational classification, not classification as such. State of Haryana v. Charanjit Singh, decided in 2006, is the authority most often cited for the restraining proposition, that parity is not to be decreed merely because two groups are employed by the same State and their work bears a family resemblance, and Jagjit Singh itself, for all its generosity to temporary employees, was careful to rest the entitlement on identity or similarity of duties and to pitch the remedy at the minimum of the regular scale rather than at full equivalence of every allowance. Read together, the constitutional cases and the statute ask the same question and accept the same defeats. Both demand a comparator; both make the comparison turn on the content of the work; both allow the difference that can be explained by a factor other than the prohibited one; and neither reaches the case in which the inequality is distributed across different jobs. The convergence is not a coincidence of drafting. It is what happens when a legal system commits itself to equality as sameness of treatment between comparable persons, a commitment that is administrable, defensible and, for the segregated workforce, systematically beside the point.

One more channel of equalisation deserves mention precisely because it was neither the statute nor the courts. The great bulk of formal-sector pay in India is fixed by the State itself, as employer, through the periodic Pay Commissions, and by industry-level settlements in banking, insurance, coal and steel. Within those systems, sex-differentiated scales had largely disappeared as a formal matter well before the Act’s machinery touched them, because a single scale for a cadre is the path of administrative least resistance once the principle of a common scale is conceded. The equal pay that most organised-sector women actually received arrived by that route, as an incident of scale rationalisation, and the Act’s role in it was atmospheric and occasionally forensic, as in Mackinnon Mackenzie, where a settlement’s fitment terms were the residue the rationalisation had missed. This is the context in which the Act’s thin docket should be read. The statute was enacted into a formal sector that was already, for its own administrative reasons, moving toward common scales for common work, and into an informal economy it could not reach. Its marginal contribution was concentrated in the contested middle, the settlements, classifications and fitments where the formal commitment to equality and the inherited structure of women’s work collided, and it is there, rather than in the aggregate statistics, that its record should be marked.

The record: a persistent gap and a quiet statute

The gap the Act was passed to close did not close. The International Labour Organization’s India Wage Report of 2018 traced the gender wage gap across two decades of household survey data and found it narrowing but stubborn, from 48 per cent in 1993-94 to 34 per cent in 2011-12, and the same organisation’s Global Wage Report 2018/19 put India’s mean gender wage gap at 34.5 per cent, the widest among the seventy three countries its hourly wage comparison covered. Figures of this kind measure the whole distance between men’s and women’s earnings, which is a much larger thing than the distance the Act policed: the gap aggregates differences in occupation, industry, hours, continuity of employment, education, the division of unpaid care work, and outright discrimination in rates, and no honest account can assign the whole of it to the last item. But the direction of the evidence is not in doubt, and neither is the statute’s place in it. A law whose claims, complaints and prosecutions were, across its working life, few enough to be individually remarkable was not a significant instrument in moving a number of that size, and the returns of the labour administration under the Act record a trickle of enforcement beside the scale of the phenomenon the Act addressed.

The reasons are structural, and they can be ranked. The first has been the subject of this article: the test required a comparator the labour market usually withholds. The second is the composition of women’s work. The Act’s machinery presupposed an establishment, an employer, prescribed registers and an identifiable workforce, and it arrived by notification in the organised employments first; the great mass of women workers were in agriculture, in casual and home-based work, and in the unorganised sector, where the establishment is diffuse, the employer may be a chain of contractors, and the male comparator, if he exists at all, is engaged by someone else. The third is the remedy’s arithmetic: a successful claim recovered the difference and nothing beyond it, so the rational claimant, weighing years of proceedings and the risk to her employment against arrears she could compute to the rupee, often had no economic case for claiming, and the rational employer, weighing the same arithmetic from the other side, had no economic case for pre-emptive compliance. The fourth is information: a claim under Section 7 depended on knowing what the men in the establishment were paid, and pay in most establishments is the least public fact in it. A claimant’s file, pay slips, the appointment letter, the settlement or award that fixed the scales, the transfer and promotion orders that showed the pattern of service, had to be assembled and kept before it could be deployed, the same discipline that leads those who handle sensitive evidence to keep it in an encrypted private file of the kind VaultBook describes for journalists protecting their sources rather than scattered across papers and inboxes that an employer’s discovery or a lost folder can undo. The fifth is the hiring margin. Part of the gap is produced before any rate is set, where an employer prices into a woman’s offered wage, or into the decision not to hire her at all, the expected cost of employing a woman, of which the largest single statutory component is maternity benefit, a cost which the Maternity Benefit (Amendment) Act 2017 substantially increased by extending paid maternity leave to twenty six weeks while leaving the cost resting on the individual employer rather than pooling it through social insurance. A remuneration statute cannot reach that calculation, which is one reason the framers paired the pay duty with a recruitment prohibition, and one reason the pairing was not enough.

The composition point deserves one further turn, because it explains why the gap statistics and the statute could pass each other for fifty years without either being mismeasured. A wage gap of the kind the ILO reports is computed over wage employees, and even within that population the largest part of the distance is a distance between jobs and sectors rather than between rates in a register: women are concentrated in agriculture and in the lower-paid services, in casual rather than regular wage work, in shorter paid hours, and in the grades of every industry from which the ladders to supervision are least often climbed. Behind the wage population stands the larger fact that a very substantial share of working women in India are self-employed or engaged as unpaid workers in household enterprises, categories in which there is no employer, no rate fixed by another party, and no comparator in any establishment, so that the entire apparatus of remuneration law, the 1976 Act and the Code alike, has no purchase on their earnings at all, however depressed those earnings are by the same social facts that depress wages. The ILO’s own analysis in the Global Wage Report made the complementary point from inside the wage population, that the gap persists among workers of comparable education and within the same occupational categories, and is wider at the top of the distribution than at the bottom, which is the shape a gap takes when segregation within occupations, into functions, grades and firms of different pay, does the work that overt rate discrimination does elsewhere. None of this acquits the statute: a law is to be judged against the harm it was enacted to address, and the harm, as the Towards Equality report described it in 1974 and as every subsequent measurement has confirmed, was always principally a harm of structure. It does, however, acquit the statistics of the charge sometimes laid against them in this debate, that a persistent gap proves the statute was simply unenforced. The gap and the statute were measuring and governing different things, and the distance between those things is the size of the definitional choice this article has been about.

None of this requires the conclusion that the Act was worthless. It established, for the first time in Indian statute law, that a sex-differentiated rate for the same work was unlawful rather than merely regrettable; it overrode settlements and contracts that embodied such rates; it gave a named authority the question of similarity to decide; and in the one case that reached the Supreme Court on clean facts it was applied with full rigour against a negotiated settlement. Statutes also work as declarations against which other institutions orient themselves, and the constitutional doctrine of equal pay, which did real work for public employees, drew part of its confidence from the existence of a parliamentary statement of the same principle. The honest verdict is narrower and harder: the Act governed the part of the pay gap that was easiest to see and least representative of the whole, and it governed that part competently. The rest of the gap lay in occupational segregation, in the structure of women’s employment and in the causes that produce both, and the Act’s instruments for those, an advisory committee and a recruitment ban keyed to the same comparator, were not built to the scale of the problem they named.

The equal value question, argued both ways

The reform proposal that follows from the diagnosis is to enact the standard India ratified: equal remuneration for work of equal value. Its strongest form runs as follows. The Convention’s Article 2 standard is not an aspiration but an obligation India accepted in 1958, and the domestic statute has never implemented it, so the gap is a deficit of compliance and not merely of policy. The standard is the only one that reaches occupational segregation, which the evidence identifies as the largest single component of the pay gap: so long as the law can only compare a woman with a man doing her job, the law structurally cannot see the mechanism by which women’s work as a class is undervalued relative to men’s work as a class, and every other defect, penalties, machinery, awareness, is secondary to that blindness. The Supreme Court in Mackinnon Mackenzie treated the Convention and European practice as the relevant horizon for the subject even while deciding on the narrower text, and the World Bank’s Women, Business and the Law assessment of India continues to identify the absence of a legal mandate of equal remuneration for work of equal value as a defect in the pay framework, which means the point is not a historical curiosity but a standing item in the external audit of Indian labour law. On this side of the argument, adopting equal value is the completion of a promise made in 1958 and renewed in 1976, and the objection that it is difficult is an objection about administration, to be answered by building the evaluation machinery rather than by declining the standard.

The case against, in its strongest form, is not a defence of unequal pay and should not be caricatured as one. Comparing dissimilar jobs requires a method: a job evaluation scheme that assigns weights to skill, effort, responsibility and working conditions across unlike occupations, and such schemes are neither neutral nor cheap. The weights embed judgments about what a job is worth, and a scheme designed by the employer whose rates are under challenge, or by consultants he appoints, can reproduce inside its own scoring the undervaluation it was commissioned to detect, so that the litigation moves from the pay packet to the architecture of the evaluation without the worker being any nearer a remedy. Jurisdictions that adopted equal value generated substantial and prolonged litigation on exactly this terrain; the British experience after equal value claims were admitted in 1983 was of claims consuming years in tribunals and appellate courts, with the European litigation over speech therapists paid less than pharmacists and other predominantly male comparators becoming the standing illustration of how a principled standard can still produce a decade of procedure before it produces a rupee, or a pound, of correction. There is also a genuine question of institutional competence: a Labour Officer deciding whether two works are of a similar nature is performing a bounded factual task, while an authority asked to rate the value of unlike jobs across an economy is performing an open-ended distributive one, and a State that has not staffed the first task convincingly has not obviously earned the second. The Indian legislature’s own conduct is evidence on this side too, and it is entitled to weight: Parliament revisited the statute in 1987 and rebuilt the whole of labour law in 2019, and on neither occasion, with the Convention’s text available to it throughout, did it adopt the wider standard.

The verdict this article defends is that the first case is right about the diagnosis and the second is right about the machinery, and that the correct reform conclusion follows from holding both. The same or similar work standard cannot reach the main body of the pay gap, and no improvement in enforcement, penalties or awareness can cure a defect of comparators; a statute book that continues to promise equal remuneration while defining the promise so that segregation defeats it is making a claim about itself that its own text refutes. But equal value enacted as a bare phrase, without a statutory job evaluation framework, an independent body to apply it, and a burden-shifting rule that puts the justification on the employer once a prima facie disparity between predominantly female and predominantly male jobs of plausibly equal demand is shown, would reproduce in India the British pattern of principled litigation and deferred remedy. The machinery question can be made concrete, because equal value regimes elsewhere have had to answer it in practice. A job evaluation scheme of the analytical kind rates jobs factor by factor: the knowledge and skill the job demands, the mental and physical effort it exacts, the responsibility it carries for people, money, equipment and outcomes, and the conditions under which it is performed, with each factor weighted and the total placing the job in a grade. The exercise sounds mechanical and is not. Deciding that responsibility for a production line’s output counts for more than responsibility for the accuracy of a label on a medicine carton, or that the physical effort of lifting counts for more than the sustained dexterity of packing, is a judgment about value dressed as a measurement, and the history of such schemes, including schemes in public employment in India drawn up for Pay Commission purposes, is that the factors in which male-dominated jobs score heavily are the factors most readily recognised as skill, while the factors in which female-dominated jobs score heavily, dexterity, care, patience, the management of interruption, are the factors most readily classified as natural attributes of the worker rather than demands of the job, and therefore scored at nothing. This is the deepest version of the case against a careless equal value reform: not that comparison is impossible, but that the comparison instrument must itself be audited for the bias it was built to detect, and that the audit requires an institution with the independence and the expertise to conduct it, which is a larger thing to build than a section in an Act. It is equally the deepest version of the case against the status quo, since the existing arrangement performs the same valuation implicitly, through the market’s inherited classifications, with no audit at all. Between an unaudited implicit valuation and an audited explicit one, the argument is not really about whether jobs will be valued across occupational lines; they already are, every time a wage board fixes relative rates. It is about whether the valuation will be visible, reasoned and challengeable, or will continue to present itself as the natural order of the pay bill.

The reform worth having is therefore institutional rather than rhetorical: the standard, plus the machinery the standard needs. It should also be said plainly, because the brief of this article requires the concession, that even that reform would address only part of the gap. Differences in hours, in continuity of service interrupted by care, in the distribution of women across industries of different productivity, and in bargaining power would survive any pay statute, and claiming the whole 34.5 per cent for Section 2(h) would overstate what a definition can do. The definition accounts for the law’s failure. It does not account for the whole of the economy’s.

The Code on Wages: where the obligation lives after the codes

The Equal Remuneration Act no longer stands on the statute book as a working law. The Code on Wages 2019, which consolidated the four wage statutes, repealed it by Section 69 of the Code along with the Payment of Wages Act 1936, the Minimum Wages Act 1948 and the Payment of Bonus Act 1965, with savings that deem action taken under the repealed Acts to have been taken under the corresponding provisions of the Code so far as not inconsistent with it. The repeal took effect when the Code was brought into force with the other three labour codes on 21 November 2025, after the long interval between enactment and commencement that the study of the labour codes implementation delay examines in its own right; the Code on Wages (Central) Rules followed on 8 May 2026, while state rules remained incomplete in much of the country, so the equal pay obligation passed, on commencement, from a statute administered employment by employment into a code whose subordinate legislation is still being assembled beneath it.

Section 3 of the Code carries the obligation, and its text deserves the same close reading as Section 4 of the old Act. It provides that there shall be no discrimination in an establishment or any unit thereof among employees on the ground of gender in matters relating to wages by the same employer, in respect of the same work or work of a similar nature done by any employee; that no employer shall, for the purpose of complying with that prohibition, reduce the rate of wages of any employee; and that no employer shall make any discrimination on the ground of sex while recruiting any employee for the same work or work of a similar nature and in the conditions of employment, except where the employment of women in such work is prohibited or restricted by or under any law for the time being in force. Three observations follow. The first is the widening the brief of this article flagged: the operative prohibition in Section 3(1) is framed on the ground of gender, a characteristic wider than the sex of the 1976 Act, extending the protection beyond the male and female binary the old definitions assumed, although the recruitment limb in Section 3(2) retains the older word sex, an inconsistency of drafting inside a single section which a court will one day be asked to reconcile and which is best read as the wider word governing the duty it introduces. The second is continuity of substance: the no-reduction rule of Section 4(2) is carried across verbatim in effect, the recruitment and conditions-of-employment prohibition of the amended Section 5 is carried across with it, and the exception for work from which women’s employment is prohibited or restricted by law survives unchanged. The third is the continuity that decides the argument of this article: the standard is still the same work or work of a similar nature, and Section 4 of the Code, which provides that a dispute whether work is of the same or a similar nature shall be decided by such authority as the appropriate Government may notify, reproduces the function of Section 7(3) of the old Act. The Code contains no equivalent of the Section 16 declaration power and no advisory committees on the employment of women, so two of the old Act’s distinctive devices disappeared in the consolidation, one of them a safety valve whose loss tightens the law and the other a promotional institution whose loss removes the only organ charged with looking at women’s employment as a whole.

A fourth observation concerns the base of the comparison, and it is the kind of change consolidation makes without announcement. The old Act compared remuneration, its own defined term, broad enough to take in the basic wage and every additional emolument in cash or in kind. The Code prohibits discrimination in matters relating to wages, and wages in the Code is a term of art with an elaborate statutory definition, built on inclusions and a schedule of exclusions, with the further rule that where the excluded components exceed one half of total remuneration they are pulled back into wages up to that threshold. A discrimination claim under Section 3 is therefore a claim about a defined quantity whose composition the Code itself regulates, and the practical character of the comparison changes with the definition: an employer who loads a package into the excluded components compresses the wages figure for men and women alike, and a differential maintained inside the excluded elements, a discretionary allowance structure that favours the grades men occupy, presents a harder question under the Code’s words than it did under the old Act’s emoluments whatsoever. Set against this, the Code’s definition is uniform across minimum wages, payment, bonus and equal pay, so the register a Section 45 authority examines is the same register every other chapter of the Code polices, a coherence the four separate statutes never achieved among themselves. Whether the narrower, more engineered base turns out to matter in litigation is a question the Code’s early case law will answer; the structural point can be made on the texts alone, that the comparison in Indian equal pay law is no longer between everything a worker receives but between a statutory quantity, and that those who drafted the consolidation understood the base of the comparison to be part of the law’s substance, since they rebuilt it everywhere else in the Code while leaving the standard untouched.

Two smaller differences between the amended Act and the Code repay notice, because consolidation is where such differences hide. The first is in the recruitment limb’s reach. The Act’s Section 5, after 1987, named its ground: discrimination in any condition of service subsequent to recruitment, such as promotions, training or transfer, an enumeration that told an authority exactly which decisions were covered and told an employer exactly which decisions to audit. The Code’s corresponding words are discrimination in the conditions of employment, a general phrase which on its face covers the same ground and more, postings, rosters, probation, confirmation, but which has shed the examples that gave the older text its grip, so that the coverage of any particular decision will have to be established by argument under the Code where the Act supplied it by list. Consolidation characteristically trades such enumerations for general words, gaining in tidiness what it risks in certainty, and the exchange is visible here in miniature. The second difference runs the other way, in the claimant’s favour, and concerns who the duty is for. The Act of 1976 was built, in its definitions, on a binary: man and woman mean male and female human beings, and Section 4’s comparison is between workers of opposite sexes so defined. The Code’s prohibition in Section 3(1) is on the ground of gender, and although the Code does not define the word, its ordinary and legal usage in the period since the Supreme Court recognised transgender persons as a third gender in National Legal Services Authority v. Union of India, decided in 2014, is wider than the binary the old Act assumed, so that a transgender employee denied equal wages for the same work has, under the Code, a textual foothold that the Act of 1976 did not offer in terms. The widening is the one point on which the Code’s restatement is substantively broader than the duty it replaced, and it arrived, characteristically, as a consequence of a change in the law’s general vocabulary rather than as a debated amendment of the equal pay rule itself.

What the claimant gained in the move is machinery, and it is considerable. Claims under the Code lie under Section 45 to an authority not below the rank of a Gazetted Officer; a single application may cover any number of employees; the authority may direct payment of the amount due together with compensation not exceeding ten times that amount, which replaces the old Act’s bare restitution with a multiplier that changes the economics of claiming; and the limitation period is three years from the date the claim arises, with power to condone delay for sufficient cause, which supplies the feature Section 7 of the old Act never had. The penalties in Section 54 punish paying an employee less than the amount due with a fine extending to fifty thousand rupees on a first offence, and a repetition within five years with imprisonment extending to three months or a fine extending to one lakh of rupees or both, with other contraventions attracting a fine extending to twenty thousand rupees. The Code also applies across establishments generally rather than by the old employment-by-employment notification, which abolishes the coverage lottery of Section 1(3) of the 1976 Act. Set against the old Act item by item, the Code is the better instrument in every particular except the one that mattered most: it is better enforced, better remedied, better timed and broader in the characteristic it protects, and it asks the same question the Ordinance of 1975 asked, whether a man and a woman in the same establishment are doing the same work or work of a similar nature. The obligation has a new home. The test moved house with it.

The transition itself deserves a paragraph, because repeals of labour statutes generate a class of practical questions that the savings clause answers only in principle. Proceedings pending before a Section 7 authority on the date of commencement, appointments of inspectors and authorities, notifications extending the old Act and declarations under Section 16 all had to find their footing in the Code, and Section 69’s saving, which deems things done under the repealed Acts to have been done under the corresponding provisions of the Code so far as they are not inconsistent with it, is the bridge on which they cross. The bridge has uneven planks. A claim for the wage difference for a period when the old Act governed is a claim about a repealed duty, maintainable through the saving and the general law of repeal, but the forum, the limitation and the compensation are the Code’s, three years and ten times under Section 45, which will in many cases be more favourable to the claimant than the regime under which her cause of action accrued; a Section 16 declaration, by contrast, has no corresponding provision in the Code to be deemed under, and its protective effect for an establishment’s differential is the kind of question the transition will produce in litigation rather than in commentary. The enforcement officer also changed character in the crossing: the Code’s Section 51 recasts the inspector as an Inspector-cum-Facilitator, appointed under an inspection scheme that may be web-based and randomised, and charged with advising employers and employees on compliance alongside investigating it, a redesign whose bearing on a duty like equal pay, where the contravention is a structure rather than an incident, remains to be worked out in practice. What can be said without speculation is that the Code assembled, for the first time, the full set of features whose absence this article has recorded in the old Act one by one: universal coverage without notification, a limitation period measured in years, compensation with a multiplier, penalties in contemporary figures and a claims authority of gazetted rank. The assembly makes the surviving defect easier to see, because it is the only one left. The codes did not fail to finish the work of 1976. They finished all of it except the definition.

One practical consequence of the transition should be recorded for the worker or adviser who arrives at this article with a live grievance rather than a historical question. The duty that governs a pay packet is the Code’s, and the claim lies under the Code’s machinery, but the subordinate apparatus through which that machinery is reached, the authorities notified for a district, the rules governing the manner of application, the inspection scheme in operation, is complete in the central sphere, where the Code on Wages (Central) Rules were notified on 8 May 2026, and remains incomplete in much of the state sphere, where State Governments have been slower to frame and notify their rules and to constitute the authorities the Code presupposes. A claimant’s first task is therefore jurisdictional in the old Act’s sense as much as the new one’s: to identify the appropriate Government for the employment, the authority notified under the Code for it, and the rules in force for the proceeding, because the Code’s text is national and its administration is not. The old Act’s claimant faced the same geography through a different door, the notification that had or had not extended the Act to her employment and the Labour Officer who had or had not been appointed for her district, and the continuity is instructive. Across fifty years and a complete rebuilding of the statute book, the distance between the equal pay duty as written and the equal pay duty as reachable has been measured, at every stage, in notifications, appointments and rules, the unglamorous instruments by which Indian labour law decides whether a right exists on the ground where the worker stands.

Verdict

The Equal Remuneration Act 1976 was not betrayed by its administrators; it was bounded by its definition. Parliament in 1975 and 1976 enacted a genuine duty, gave it overriding effect against the settlements that embodied the old differentials, forbade compliance by levelling down, and in 1987 widened its discrimination ban across the whole of a woman’s service and armed it with minimum imprisonment. What Parliament never did, in 1976, in 1987 or in the consolidation of 2019, was enact the standard India had accepted at Geneva in 1958, and without that standard the statute could only ever police the minority of pay inequality that presents itself as two workers, two sexes, one job and two rates. The reader who understands Section 2(h) understands the Act: why its leading case is a single fitment dispute from a Bombay shipping office, why the celebrated equal pay doctrine grew up in constitutional litigation about temporary and permanent government servants instead, why the gap statistics barely moved in five decades of the statute’s existence, and why the Code on Wages, for all its superior machinery, inherited the original limitation intact. The standard is the whole problem, and the standard is a choice, which means it can be chosen differently.

Frequently Asked Questions

Q: What did the Equal Remuneration Act 1976 require?

The Act imposed two duties on every employer in an establishment or employment to which it had been extended. Under Section 4, the employer had to pay remuneration at rates not less favourable to workers of one sex than to workers of the opposite sex for performing the same work or work of a similar nature, without reducing any worker’s rate in order to comply, and pre-existing differentials based only on sex were levelled up to the higher or highest rate from commencement. Under Section 5, the employer could not discriminate against women in recruitment for such work, and after the amendment of 1987 could not discriminate in conditions of service after recruitment either, such as promotions, training or transfer, except where women’s employment in the work was prohibited or restricted by law.

Q: What is same work or work of a similar nature?

Section 2(h) of the Act defines it as work in respect of which the skill, effort and responsibility required are the same, when performed under similar working conditions, by a man and a woman, where the differences, if any, between the skill, effort and responsibility required of a man and those required of a woman are not of practical importance in relation to the terms and conditions of employment. The definition therefore requires a comparison between the work of a man and the work of a woman in the same establishment. Where a question arose whether two works were of the same or a similar nature, Section 7(3) gave the question to the authority appointed under the Act to decide, and Section 4 of the Code on Wages assigns the same question to a notified authority.

Q: Does Indian law require equal pay for work of equal value?

No. Indian law has never enacted the equal value standard. The Equal Remuneration Act 1976 required equal pay for the same work or work of a similar nature, and the Code on Wages 2019 retained that test in Section 3. Equal remuneration for work of equal value is the standard of Article 2 of the International Labour Organization’s Equal Remuneration Convention, 1951 (No. 100), which India ratified on 25 September 1958; it permits a comparison between two different jobs rated equal in value on skill, effort, responsibility and working conditions. Because occupational segregation means women frequently have no male comparator doing the same or similar work, the narrower domestic standard leaves the largest part of the pay gap outside the law’s reach.

Q: Can an employer justify a pay difference between men and women?

A pay difference is not a contravention merely because it exists; it must rest on the ground of sex. Differences explained by a factor other than sex, such as seniority, qualifications, merit, the quality or quantity of output, or the genuinely different content of a job, are consistent with the Act, and an employer could prove such a factor before the authority hearing a claim. The justification fails where the supposed factor is the discrimination in another dress, as where a job title or a settlement’s fitment terms reserve the better treatment to men. The Act also allowed the appropriate Government, under Section 16, to declare a difference in an establishment to be based on a factor other than sex, in which case acts attributable to it were deemed not to be a contravention.

Q: How does a worker enforce an equal pay claim?

Under the Act, the worker applied to the authority appointed under Section 7 for the local area, an officer not below the rank of a Labour Officer, who could hear both complaints of contravention and claims for non-payment of wages at equal rates. The authority had the powers of a civil court for evidence and attendance, decided any question whether the works were of the same or a similar nature, and could direct payment of the difference between the wages payable and the wages paid, recoverable through Section 33C of the Industrial Disputes Act 1947. An appeal lay within thirty days. A prosecution was also possible, and after 1987 the person aggrieved or a recognised welfare organisation could file the complaint. Claims arising after the repeal go to the authority under Section 45 of the Code on Wages.

Q: What replaced the Equal Remuneration Act?

The Code on Wages, 2019 replaced it. Section 69 of the Code repealed the Equal Remuneration Act 1976 together with the Payment of Wages Act 1936, the Minimum Wages Act 1948 and the Payment of Bonus Act 1965, with savings for action taken under the repealed Acts. The repeal took effect when the four labour codes were brought into force on 21 November 2025. The equal pay obligation itself continues in Section 3 of the Code, which prohibits discrimination on the ground of gender in wages for the same work or work of a similar nature, carries forward the ban on reducing wages to comply, and restates the recruitment and conditions-of-employment prohibition, with disputes about similarity decided under Section 4 of the Code.

Q: Did the Code on Wages widen the equal pay protection?

In part. Section 3(1) of the Code frames the prohibition on the ground of gender, a wider characteristic than the sex of the 1976 Act, whose definitions assumed a male and female binary; and the Code applies generally rather than through the employment-by-employment notifications the old Act required, while its claims chapter gives a three-year limitation period and compensation of up to ten times the amount due, neither of which the old Act had. What the Code did not widen is the standard of comparison: the duty still attaches to the same work or work of a similar nature, so the equal value standard of the Convention India ratified in 1958 remains unenacted. The Code also dropped the old Act’s advisory committees and its Section 16 declaration power.

Q: Why has the equal pay law achieved so little?

The principal reason is the standard the law adopted. A duty keyed to the same work or work of a similar nature needs a comparator of the opposite sex doing that work in the same establishment, and occupational segregation, the sorting of women into different jobs, processes and sectors from men, usually removes the comparator before any question of enforcement arises. The supporting causes compounded the defect: the Act reached employments only by staged notification and arrived in agriculture only in 1978; a successful claim recovered the wage difference and nothing more, which priced claiming above its return for most workers; prosecutions were gated through the government until 1987 and the original fines were one thousand and five thousand rupees; and the advisory committees charged with segregation itself could only advise.

Q: What is the constitutional basis for equal pay in India?

The express constitutional statement is Article 39(d), a Directive Principle of State Policy directing the State to secure equal pay for equal work for both men and women; Directive Principles are fundamental in governance but not enforceable in court by themselves. The enforceable doctrine was built on the fundamental rights: in Randhir Singh v. Union of India, (1982) 1 SCC 618, the Supreme Court held that equal pay for equal work is deducible from Articles 14 and 16, the guarantees of equality and of equality of opportunity in public employment, construed in the light of the Preamble and Article 39(d), so that unequal pay for identical work under the same employer, resting on no rational classification, is unconstitutional. Article 15’s prohibition of discrimination on the ground of sex stands behind the structure.

Q: What penalty does the Equal Remuneration Act prescribe for paying unequal wages?

After the amendment of 1987, Section 10(2) punished the substantive offences, including making any payment of remuneration at unequal rates to men and women workers for the same work or work of a similar nature, with a fine of not less than ten thousand rupees and extending to twenty thousand rupees, or imprisonment for a term of not less than three months and extending to one year, or both, for the first offence, and with imprisonment extending to two years for the second and subsequent offences. Failures of record and cooperation under Section 10(1), such as omitting to maintain or produce registers, attracted simple imprisonment extending to one month or a fine extending to ten thousand rupees or both. Before 1987 the fines had extended only to five thousand rupees for the substantive offences and one thousand for the record offences.

Q: Can an employer cut men’s wages to comply with the equal pay duty?

No. Section 4(2) of the Act provided that no employer shall, for the purpose of complying with the equal pay duty, reduce the rate of remuneration of any worker, so equalisation had to be achieved by raising the lower rate. Section 4(3) applied the same logic to differentials that already existed when the Act commenced in an establishment: where the rates for men and women for the same work or work of a similar nature differed only on the ground of sex, the higher rate where there were two, or the highest where there were more than two, became the rate payable to both, although the proviso denied any revision for service rendered before commencement. The Code on Wages carries the no-reduction rule forward in Section 3(2).

Q: Is there a time limit for making a claim under the Equal Remuneration Act?

The Act prescribes no limitation period for making a claim or a complaint before the authority under Section 7, a point on which summaries of the Act are often wrong. The only periods the Act fixes concern the appeal: an employer or worker aggrieved by the authority’s order had thirty days from the date of the order to appeal to the notified appellate authority, and the appellate authority could allow a further thirty days where the appellant had been prevented by sufficient cause, but not thereafter. The position changed with the successor law: Section 45 of the Code on Wages fixes a limitation period of three years from the date on which a claim arises, with power in the authority to condone delay for sufficient cause.

Q: What did the Equal Remuneration (Amendment) Act 1987 change?

The Equal Remuneration (Amendment) Act, 1987 (Act 49 of 1987), in force from 16 December 1987, made four changes. It extended Section 5 beyond recruitment to any condition of service subsequent to recruitment, such as promotions, training or transfer. It substituted the penalties in Section 10, replacing fines extending to one thousand and five thousand rupees with the higher scale described in this article, including minimum terms of imprisonment for substantive offences. It substituted Section 12 so that a prosecution could begin on the complaint of the person aggrieved or of a recognised welfare institution or organisation, and not only of the government. And it substituted Section 15, the special cases provision. It did not alter Section 2(h) or the same or similar work standard.

Q: Which Supreme Court cases built the equal pay for equal work doctrine?

The doctrine is principally constitutional and concerns public employment. Randhir Singh v. Union of India, (1982) 1 SCC 618, a Delhi Police driver’s claim, derived the principle from Articles 14 and 16 read with Article 39(d). Dhirendra Chamoli v. State of Uttar Pradesh and Surinder Singh v. Engineer-in-Chief, CPWD, both decided in 1986, extended it to casual and daily-wage workers doing the same work as regular employees. State of Punjab v. Jagjit Singh, (2017) 1 SCC 148, restated the line for temporary employees, entitling them to wages at the minimum of the regular pay scale for the same duties. The leading case under the Equal Remuneration Act itself is Mackinnon Mackenzie and Co. Ltd. v. Audrey D’Costa, (1987) 2 SCC 469, the women stenographers’ case, which is a different and much smaller line of authority.

Q: Does the equal pay duty cover allowances and benefits in kind, or only basic wages?

It covers the whole package. Section 2(g) defines remuneration as the basic wage or salary, and any additional emoluments whatsoever payable, either in cash or in kind, to a person employed in respect of employment or work done in such employment, if the terms of the contract of employment, express or implied, were fulfilled. Allowances, bonuses and benefits provided in kind are therefore inside the comparison, and an employer could not satisfy the duty by equalising the basic rate while differentiating in the allowances attached to the work. The comparison under Section 4 is of rates of remuneration as so defined, which is why a pay audit under the Act had to proceed element by element across the wage structure rather than by comparing headline wages alone.

Q: What records must an employer keep under the Equal Remuneration Act?

Section 8 required every employer, on and from the commencement of the Act, to maintain such registers and other documents in relation to the workers employed by him as may be prescribed, the register under the Equal Remuneration Rules being maintained in Form D. The registers were the evidentiary base of the whole scheme: they recorded the workers and the rates paid to them, so that a differential between men and women doing the same work could be seen on the face of a document rather than reconstructed from testimony. Inspectors appointed under Section 9 could require production of the registers, muster rolls and other documents, examine them and take copies or extracts. Omitting or failing to maintain or produce them was an offence under Section 10(1), punishable with simple imprisonment extending to one month or a fine extending to ten thousand rupees or both.

Q: What do the advisory committees under the Equal Remuneration Act do?

Section 6 required the appropriate Government to constitute one or more advisory committees for the purpose of providing increasing employment opportunities for women, to advise it with regard to the extent to which women may be employed in such establishments or employments as may be specified. Each committee was to consist of not less than ten persons nominated by the appropriate Government, of whom one-half were to be women. In advising, the committee had regard to the number of women employed in the establishment or employment, the nature of the work, hours of work, the suitability of women for employment, and the need for increasing employment opportunities for women, including part-time employment. After considering the advice and hearing representations, the Government could issue directions on the employment of women workers, and failure to carry out such a direction was an offence under Section 10(2).

Q: What happened to claims begun under the Equal Remuneration Act after its repeal?

The repeal by Section 69 of the Code on Wages, 2019 was accompanied by a saving: actions taken under the repealed Acts, including notifications issued, appointments made and amounts payable, are deemed to have been taken under the corresponding provisions of the Code so far as they are not inconsistent with it. A claim about a wage difference that accrued while the old Act governed therefore does not die with the repeal, but the proceeding is carried into the Code’s framework, with its authority under Section 45, its three-year limitation and its compensation of up to ten times the amount due. Some old devices have no corresponding provision to be saved into: the Section 16 declaration power and the advisory committees disappeared in the consolidation, since the Code contains no equivalent of either.

Q: How large is the gender pay gap in India?

The figure depends on the measure and the source, and should always be taken with both attached. The International Labour Organization’s Global Wage Report 2018/19 put India’s mean gender wage gap at 34.5 per cent, the widest among the seventy three countries covered by its hourly wage comparison. The same organisation’s India Wage Report of 2018, working from household survey data, traced a narrowing over two decades, from 48 per cent in 1993-94 to 34 per cent in 2011-12. These figures measure the whole distance between men’s and women’s earnings, which aggregates occupational segregation, hours, continuity of employment and the distribution of women across industries along with discrimination in rates for the same work; they are not a measure of contraventions of the equal pay statute alone.

Q: Did the Equal Remuneration Act cover agricultural and unorganised sector workers?

Only in a qualified sense. The Act did not apply to any employment until the Central Government extended it by notification under Section 1(3), and the definition of worker in Section 2(i) was tied to establishments and employments in respect of which the Act had come into force. The notifications followed the organised sector first, hospitals in January 1976, banks and insurance in March 1976, mines and educational institutions in April 1976, and agriculture was reached only on 26 January 1978, with community, social and personal services following in June 1978. Even after extension, the Act’s machinery presupposed an identifiable employer, an establishment and prescribed registers, and the comparator the duty required was hardest to find in casual, seasonal and home-based work, so coverage on paper and coverage in practice diverged most widely in the unorganised sector.