The albatross changes necks: what the ‘industry’ verdict leaves to the new Labour Code

The Rime of the Ancient Mariner, a poem by Samuel Taylor Coleridge, written in 1798, is about how a sailor shoots an albatross—a bird believed to be a sign of good fortune. The ship’s crew then dies one by one. The survivors hang the dead bird around the mariner’s neck. He must carry it until he finds a way to atone. 

From the poem, the phrase “an albatross around one’s neck” entered the language as a description of a burden one cannot put down, usually a burden of one’s own making.

Chief Justice Surya Kant opens his opinion with that image. The definition of “industry” under Section 2(j) of the Industrial Disputes Act, 1947, he writes, has been “a proverbial albatross fastened to the Court’s metaphorical neck”. It is a question that has returned again and again despite repeated attempts to settle it. He ends his 154-page-long opinion with the hope that the judgement “finally lifts that burden and permits the law to move forward unencumbered”.

The 489-page-long judgement attempts to loosen the knot by dealing with a question that was referred 21 years ago in 2005. Five out of nine judges authored distinct opinions. Read together, they decide less than what many had expected. What the judges did, in the end, was pass the albatross on.

Why the question mattered

The 1947 Act gives workmen access to conciliation, labour courts and tribunals, and the prospect of reinstatement with back wages. None of that is available unless the employer runs an “industry”. The word is defined under Section 2(j) of the Act. It lists what an industry must include, but never says what it excludes. The answer to the question ‘who is a worker?’ may well turn on a prior one: who is an employer running an “industry”? Read literally, Justice P.S. Narasimha observes, it can describe a factory and a family kitchen in the same breath.

In 1978, a seven-judge bench in Bangalore Water Supply v A. Rajappa settled the matter, or appeared to. Justice V.R. Krishna Iyer’s “triple test” asked three questions. 

1. Is there systematic activity? 
2. Is it organised through cooperation between employer and employee? 
3. Is it for the production or distribution of goods or services to satisfy human wants?

If yes, the enterprise is an industry. The profit motive was irrelevant. Hospitals, universities, municipal boards, clubs and charities came within the Act. Only essential or inalienable sovereign functions of the State stayed out.

The ruling was meant to be provisional. Justice Iyer concluded “with diffidence” and urged Parliament to legislate a tidier definition. Parliament did, by an amendment in 1982, and never notified it. For 48 years the interim solution held the field.

The long reference

A five-judge bench in 2005 was dealing with a question: does Uttar Pradesh’s Social Forestry Department, a welfare initiative, pass off as an “industry”? The Bench, in State of U.P. v Jai Bir Singh (2005), doubted the breadth of the triple test and sent the matter to the then Chief Justice of India to constitute a larger bench to decide the issue. On 2 January 2017, a seven-judge bench directed that nine-judges hear it. The hearing finally ran over three days in March 2026.

In the meantime, the ground had shifted. The Industrial Relations Code, 2020 came into force on 21 November 2025. The 1947 Act and Section 2(j) was repealed, replaced by a new definition in Section 2(p). While Section 2(p) borrows the language of the triple test almost verbatim, it expressly excludes charitable institutions, domestic service and activities “relatable to” sovereign functions. The nine judges were interpreting a repealed provision for the benefit of disputes filed before its repeal.

What the Chief Justice wrote

CJI Surya Kant, writing for himself and Justices Satish Chandra Sharma, Alok Aradhe and Vipul M. Pancholi, held the reference valid and answered it on merits. The triple test, he concluded, “has stood the test of time”, but some elements “could have been articulated differently”. He then set out how it “would have been more appropriately structured, hypothetically speaking”.

The reformulated triple test, as per CJI Surya Kant’s opinion, maintains the first two limbs and changes the third. The goods or services must possess “a discernible commercial character that is analogous to trade or business”. The opinion carefully separates commerce from profit. An activity may be commercial without seeking gain. But activities “wholly divorced from commercial or economic enterprise” would ordinarily fall outside the definition. The sovereign-functions exception survives, as does the dominant-nature test for mixed undertakings. A trade does not cease to be one merely because philanthropy “animates the undertaking in its official documentation”.

What would the new limb exclude? The reformulated test carves out activities “purely spiritual or religious in nature”. Material goods or services that happen to serve a religious object stay in. For context, The old operative Section 2(j) did not itself contain the spiritual-or-religious exclusion. That limitation came from Bangalore Water Supply. Parliament incorporated it into the 1982 amended definition, but that amendment was never brought into force. The 2020 Code now expressly puts the exclusion into the statutory text. More broadly, the CJI Surya Kant’s opinion says activities bearing no “meaningful analogy to trade, business, or economic enterprise” would ordinarily fall outside the definition. The Chief Justice insists this does not “denude workmen of statutory protection”. Other judges on the Bench disagree. Justice Dipankar Datta notes that a hospital technician is a workman whether the hospital runs on charity or on profit.

Then comes the qualification that defines the judgement. The reformulated test “is not intended to displace the governing legal position in respect of pending proceedings”. Every matter pending under the 1947 Act may be decided under the 1978 test. Finalised awards stand undisturbed. The Social Forestry appeals, referred in 2005 to test the 1978 ruling, will now be decided under it. The bench “refrain[s] from entering into an examination” of the Code’s definition. It adds that Bangalore Water Supply should not be the Code’s “sheet anchor” either.

Counting the votes

Early reports described a 5:4 majority for the reformulation. The signatures do not support that count. Four judges signed the Chief Justice’s opinion. Justice Narasimha, writing separately, agreed the reference was valid but held that an authoritative interpretation “has now become unnecessary” after repeal. Modifying the ratio, he warned, would create “a third category” of law applicable only to pending cases. A narrower test would push long-pending disputants “to file fresh suits before the Civil Court”. Justice Joymalya Bagchi also upheld the reference, but found the triple test “exhibits a rare felicity in judicial interpretation”. Economic and technological change since 1978, he added, had “weakened rather than strengthened” the case for revisiting it.

Justice B.V. Nagarathna held the reference “uncalled for” and gave ten reasons why it need not be answered. One was that the reforms of 1991 had privatised much of what the State once did, which counselled stability over revision. Justices Datta and Ujjal Bhuyan, in a joint opinion, rejected the reference outright. “Institutional credibility lies in respecting finality, not in perpetuating doubt,” Justice Datta wrote.

The arithmetic, then, is this: Six judges found the reference valid. Four reformulated the test. Five declined to touch Bangalore Water Supply. All nine agreed that pending cases remain governed by 1978, and that the new Code must be read on its own terms.

Justice Nagarathna agreed that Section 2(p) of the 2020 Code must be interpreted on its own terms, but clarified that Bangalore Water Supply could still be relevant where the new and repealed definitions share commonalities or similarities, and rejected the suggestion that the precedent must be given a complete “go-by.”

What the ruling does

A sceptic can put the point sharply. Four judges have rewritten a test and have declared that it governs no pending case. They have declared that it does not govern the provision that replaced it. CJI Surya Kant’s own adverb for the exercise is “hypothetically”. On that reading, 489 pages change nothing.

The political criticism has been explicit. Jairam Ramesh, the Congress general secretary in charge of communications, argued on X (formerly Twitter) that the ‘majority’ narrows the 1978 approach in two ways. It demands a commercial character resembling trade or business, a requirement absent from the earlier test. And it widens the sovereign functions exemption, potentially placing more government activity outside the definition. Leaving a hypothetical reformulation on the record, he said, creates a gap that invites litigation before labour courts and tribunals. He pointed out that the Code already empowers the Union Government to exclude further categories of establishments. He praised Justice Nagarathna’s dissent as courageous and clear. Any move to narrow the definition, he said, “risks weakening worker safeguards precisely when they are needed most”.

Trade unions have gone further. Amarjeet Kaur, general secretary of the All India Trade Union Congress, said the ruling could particularly affect workers in hospitals and educational institutions. It could weaken unionisation in those sectors, she added. C. Srikumar, the union’s national secretary, called the reformulation a cure “worse than the illness”.

The sceptic will reply that all of this criticism is aimed at a test with no application.

That reading understates the judgement’s institutional weight. A majority view confirmed that a smaller bench may, through the CJI, send a larger bench’s ruling for reconsideration, as Dawoodi Bohra permits. Nine judges have told tribunals exactly how to dispose of legacy cases, a real relief to litigants who have waited two decades. And the CJI and three colleagues have reasoned that “industry” cannot be “wholly divorced from its commercial character”. That passage will be cited the day Section 2(p) of the new Code reaches the Court. Disclaimers in a judgement travel less well than its reasoning.

Justice Narasimha saw the risk and named it. The Code, he wrote, must be “freed of the bondage” of the old ratio and of “any new definitions that we may propose”. He was right to worry. The 2020 Code defines “industry” afresh, and every dispute arising after 21 November 2025 will be decided under that definition, not the old one. Sooner or later a labour court will have to decide whether a hospital, a university or a government welfare scheme is an “industry” under the new Code. Counsel on one side will cite the CJI’s reasoning on commercial character. Counsel on the other will cite the disclaimer that the reasoning was “hypothetical”—never meant to govern the Code. Whether a future bench treats the commercial-character limb as persuasive authority or as expressly disclaimed is the question this judgement leaves open.

In Coleridge’s poem, the mariner is freed from the albatross only when he learns to bless the creatures of the sea, and even then he is condemned to retell his story to every stranger he meets. The Court has not been freed. It has retold the story, at great length, and the bird has simply moved. For 48 years it hung on Section 2(j) of a 1947 Act. It now hangs on Section 2(p) of a 2020 Code, waiting for the next reference.