What Is an 'Industry' in Indian Labour Law? The Bangalore Water Supply Test and Its Limits

The Industrial Disputes Act, 1947 governs strikes, lockouts, layoffs, retrenchment, closures, and the compulsory adjudication of collective disputes between workers and employers. The Industrial Relations Code, 2020, which replaced it with effect from 21 November 2025, carries the same essential architecture forward. Both statutes apply only to “industries.” Whether a particular enterprise — a government hospital, a university, a municipality, a charitable trust, a research institute, a law firm, or a tech company’s back-office operation — is an “industry” determines whether its workers can raise industrial disputes, whether they are protected from arbitrary retrenchment, whether they have access to the conciliation, arbitration, and adjudication machinery of the statute, and whether the employer requires government permission before effecting large-scale layoffs or closures.

This is not abstract jurisprudence. It is the threshold question in every matter that comes through the doors of an industrial tribunal or a labour court. Get it wrong, and entire chapters of statutory protection either appear or disappear depending on which direction you err. A senior practitioner advising on an acquisition, a restructuring, or a dispute strategy must resolve it before any other question can be answered.

Section 2(j) contains the original definition. It is deceptively short. It has generated more litigation in Indian labour law than almost any other provision, and it has been interpreted, contested, constitutionally re-examined, legislatively addressed, and then addressed again in a codified form. The starting point is the statutory text. The endpoint, for current purposes, is the Industrial Relations Code, 2020. Between them lies one of the most analytically substantial bodies of doctrine in Indian labour law.

I. The Statutory Text — Section 2(j) of the Industrial Disputes Act, 1947

Section 2(j) defines “industry” as:

any business, trade, undertaking, manufacture or calling of employers and includes any calling, service, employment, handicraft, or industrial occupation or avocation of workmen.¹

The definition is constructed in two limbs. The first — “business, trade, undertaking, manufacture or calling of employers” — approaches the question from the employer’s perspective. The second — “calling, service, employment, handicraft, or industrial occupation or avocation of workmen” — approaches it from the workmen’s perspective.

The structure is unusual and its implications are significant. Most definitional provisions in labour legislation take a single axis. Section 2(j) takes two, and the inclusive language (“includes any calling, service, employment”) signals expansiveness rather than limitation.

The word “undertaking” in the first limb carries no inherent commercial connotation. A government department could be an undertaking. A public hospital could be an undertaking. A university could be an undertaking. Courts have consistently held that “undertaking” connotes any organised systematic activity involving the deployment of labour, material, and resources toward a defined objective — commercial or otherwise. The absence of a profit motive does not on the face of the text disqualify an enterprise.¹

The second limb’s inclusion of “service” is equally expansive. A professor at a university renders a service. A doctor at a charitable hospital renders a service. A security guard at a religious institution renders a service. Does the inclusion of “service” bring all of these within the Act? The courts held that the two limbs must be read together, purposively, against the objectives of the statute. But that purposive reading produced conflicting results for thirty years until a seven-judge Constitution Bench intervened.

II. The Doctrinal Development Before Bangalore Water Supply

The definitional controversy has three phases before 1978, each producing a different analytical approach.

The earliest position was commercially centred. Some decisions in the 1940s and 1950s held that an activity could only be an “industry” if the employer was engaged in business or trade with a commercial character. On this view, government departments, charitable hospitals, and educational institutions could not be industries because the employer lacked a profit motive. This approach was narrower than the statutory text warranted and was progressively abandoned.²

The critical reorientation came with D.N. Banerji v. P.R. Mukherjee AIR 1953 SC 58, in which the Supreme Court held that a municipality was an industry under the Act.³ The Court focused on the organised character of the activity and the cooperation between employer and workmen — not on commerciality. The employer’s purpose was held to be irrelevant; what mattered was the nature and structure of the enterprise.

State of Bombay v. Hospital Mazdoor Sabha AIR 1960 SC 610 extended the analysis to hospitals.⁴ The Court applied what later came to be called the dominant-nature approach: even if an institution had multiple activities, the question was what its dominant character was. If the dominant activity involved the production or delivery of goods and services through organised employer-employee cooperation, the institution was an industry regardless of its formal purpose or ownership structure.

By the mid-1960s, courts were applying a functional test: does the enterprise have the structural attributes of industry — systematic organisation, cooperation between employer and worker, activity directed toward a result? If yes, it could be an industry without commerciality. This created considerable uncertainty. Different High Courts applied the test differently. Hospital employees, university staff, municipal workers, and employees of welfare organisations won or lost depending on which formulation a particular bench preferred. By the time the issue reached the Constitution Bench, the law was fragmented.²

III. Bangalore Water Supply and Sewage Board v. A. Rajappa (1978) — The Seven-Judge Bench

The definitive answer came from a seven-judge Constitution Bench in Bangalore Water Supply and Sewage Board v. A. Rajappa (1978) 2 SCC 213.⁵ Justice V.R. Krishna Iyer wrote the leading judgment. It is the most cited authority in Indian labour law — a practitioner’s essential reference, applied in hundreds of reported decisions across every level of the judicial hierarchy.

The Triple Test

The Court synthesised the entire prior jurisprudence into a three-part conjunctive test. An entity is an “industry” under Section 2(j) if:

First, there is a systematic or organised activity. Casual or isolated activities do not qualify. The word “systematic” refers to an enterprise that operates with regularity, structure, and continuity — not to a one-off event or an occasional transaction. A municipality providing water supply across an entire city satisfies this; an individual hiring a carpenter for a single repair does not.

Second, there is cooperation between an employer and employees. This element excludes purely individual work — a sole practitioner with no employees is not running an “industry” in the statutory sense. The enterprise must involve an employment relationship, with an employer directing the activity and workers executing it.

Third, the activity results in the production or distribution of goods or the rendering of services. The services may be material or non-material, and they may be provided for payment or without charge. Education imparts a service of knowledge. Medical care delivers a service of health. Sewage treatment distributes a service of sanitation. None of these requires a commercial transaction.⁵

If all three elements are present, the enterprise is an “industry” regardless of whether it is run for profit or without profit, whether it is government-owned or privately owned, whether its purpose is charitable, educational, religious, or commercial.

The Dominant Nature Test

For enterprises that carry out multiple activities, some of which would be industrial and some of which would not, the Court formulated the dominant nature test: the question is what the dominant character of the enterprise is. If the dominant activity is industrial in character, the whole enterprise falls within Section 2(j). If only a severable and identifiable part of the enterprise is industrial, that part alone is an “industry” and the remaining functions are excluded.⁵

This test requires a fact-specific inquiry. A university with both teaching functions and maintenance, security, and administrative staff requires an assessment of whether the dominant activity — education — satisfies the triple test, and if not, whether the non-teaching operations are sufficiently severable to be treated independently.

What the Court Included

Applying the triple test, the Court held that the following are industries: municipalities and local bodies, sewage and water boards, charitable and government hospitals, educational institutions at least in respect of non-teaching staff, research institutes with organised technical and scientific staff, clubs with paid employees, cooperatives, and professional associations with organised workforces.⁵

The judgment expressly held that the profit motive is irrelevant to the classification. This was the most radical holding, and it had immediate and enormous practical consequences. A charitable hospital that charges no fees is an industry if it employs doctors, nurses, and ancillary staff in an organised way and renders medical services. The purpose of the employer — whether to earn profit, to serve the community, to fulfil a religious obligation, or to discharge a constitutional function — does not change the structural character of the employment relationship within the enterprise.

This holding extended coverage to government enterprises, public sector undertakings, municipal corporations, universities, and the full range of institutions that the welfare state had created. The workers employed in all of these entities could now raise industrial disputes, claim retrenchment compensation, and access the conciliation and adjudication machinery of the Act.

What the Court Excluded

The exclusions the Court recognised were narrow. Purely sovereign functions of the State — the armed forces, police in their law-enforcement capacity, the legislature, and the core administrative functions of taxation and revenue collection — were excluded because the employer-employee relationship in those contexts is not analogous to the industrial relationship.⁵ A customs officer exercising sovereign authority is not in the same structural position as a factory worker subject to an employer’s economic power.

Individual practitioners — a sole medical practitioner, a solo lawyer, an individual consultant with no employees — were excluded because there is no organised cooperation between employer and employees.

Purely domestic service — workers employed in individual private households — was excluded as outside the industrial relationship.

Everything outside these narrow categories was, broadly, included. The judgment ran to over a hundred paragraphs and included a long list of illustrative inclusions and exclusions in its closing portion — a practitioner’s checklist that has been consulted in every subsequent industry-definition dispute.

IV. The Normative Logic — Why the Broad Interpretation Is Defensible

The Bangalore Water Supply decision can be criticised for flooding the industrial dispute machinery with entities ill-suited to collective bargaining. That criticism is addressed below. But it can also be defended at a level deeper than doctrinal analysis.

The purpose of the Industrial Disputes Act is to regulate the employment relationship and protect workers from arbitrary action. If the definition of “industry” is drawn narrowly — confined to commercial, profit-making enterprises — then the workers employed by governments, hospitals, universities, and nonprofits are excluded from the statute’s protective architecture entirely. Those workers are not in any structurally different position from a factory worker. The power asymmetry between a hospital management and its cleaning staff, or between a university administration and its security guards, is identical in structure to the relationship between a factory owner and a production worker. The vulnerability to arbitrary dismissal is the same. The need for collective representation and procedural protection is the same.

To exclude these workers from the Act simply because their employer is a charitable trust rather than a limited company, or a government department rather than a private enterprise, is to allow legal form to override substantive reality. The broad reading of Section 2(j) prevents that substitution. It ensures that the statute’s protective logic reaches the employment relationship wherever that relationship exhibits the structural inequality that makes protection necessary.⁶

That is the normative case for Bangalore Water Supply. It is grounded in the same Kahn-Freund insight that runs through the labour law series: law must correct unequal bargaining power, not entrench it by drawing arbitrary distinctions between classes of employers.

V. The Criticism and the Reference — State of U.P. v. Jai Bir Singh

The Bangalore Water Supply decision was controversial from the moment it was delivered. The criticism was not that the reasoning was incoherent, but that the reach was too wide. Opponents argued that pulling government departments, pure administrative bodies, charitable trusts, and welfare organisations into the industrial dispute machinery created obligations — formal conciliation proceedings, adjudication before labour courts, mandatory retrenchment compensation, prior permission for closure — that were unsuited to the institutional character of those entities. Municipal corporations going on strike, universities locked out, government offices paralysed by industrial action — these were outcomes that critics argued the legislature never intended.⁷

The institutional argument also had force. Industrial dispute adjudication was designed for a world of factory production, where the parties — an employer and a defined workforce — are in a clear economic relationship. When the “industry” is a government department or a charitable hospital, the interests at stake are public interests as much as private ones. The adjudication machinery was not designed to balance those considerations.⁷

These arguments were formally registered at the Supreme Court level in State of U.P. v. Jai Bir Singh, where a smaller bench expressed doubt about the width of the seven-judge decision and referred it to a larger bench for reconsideration.⁷ The referral created a paradox: a smaller bench cannot overrule a larger one, but it can express doubt and refer. The referral did not suspend the operation of Bangalore Water Supply — lower courts and tribunals continued to apply the triple test — but it created uncertainty about the long-term future of the doctrine.

The nine-judge Constitution Bench constituted to hear the Jai Bir Singh reference heard final arguments over 17, 18, and 19 March 2026 and reserved judgment on 19 March 2026. As of April 2026, the judgment has not yet been delivered. Until it is, Bangalore Water Supply remains the binding authority for disputes arising under the Industrial Disputes Act, 1947. Whatever the nine-judge bench decides will reframe the doctrine — either by confirming, modifying, or narrowing the triple test — and its reasoning will govern the interpretation of the corresponding definition in the Industrial Relations Code, 2020.

For practitioners advising clients today, the practical position is to assume that Bangalore Water Supply governs IDA-era disputes while monitoring the pending judgment closely. For disputes arising under the new Code, the statutory definition and its express exclusions are the primary text, with the triple test operating as an interpretive framework where the statutory language requires construction.⁷

VI. The Legislative History — The Unnotified 1982 Amendment

There is a persistent and important error in accounts of this area that must be corrected. The Industrial Disputes (Amendment) Act, 2010 did not amend the definition of “industry” in Section 2(j). The 2010 amendment addressed different provisions — principally the definition of “appropriate Government” and the provisions on individual disputes in Section 2A. The definition of “industry” in Section 2(j) was not touched.⁸

The actual legislative response to the Bangalore Water Supply controversy was the Industrial Disputes (Amendment) Act, 1982, which substituted a new and significantly narrower definition of “industry” in Section 2(j). The 1982 amendment would have expressly excluded hospitals, educational institutions, charitable bodies, sovereign functions of government, and domestic service from the definition. It represented Parliament’s direct response to the criticism that the seven-judge bench had gone too far.

The 1982 amendment was never brought into force. It received presidential assent, it was published in the Gazette, but the notification required to commence it was never issued. The result was that the old broad Section 2(j) — as interpreted by Bangalore Water Supply — continued to operate as the law until the Industrial Relations Code, 2020 replaced it.⁸

This legislative history matters for three reasons. It shows that Parliament accepted the principle that Bangalore Water Supply had gone too far — which is relevant to how the nine-judge bench might approach the reconsideration. It explains why the Industrial Relations Code, 2020 contains explicit exclusions in its definition — the Code effectively enacts what the unnotified 1982 amendment proposed. And it demonstrates that the definitional controversy was not resolved by judicial or legislative action for nearly four decades, during which the triple test continued to apply.

VII. The Industrial Relations Code, 2020 — The Current Position

The Industrial Relations Code, 2020, which consolidated the Trade Unions Act, 1926, the Industrial Disputes Act, 1947, and the Industrial Employment (Standing Orders) Act, 1946, came into force on 21 November 2025.⁹ Section 2(p) of the Code defines “industry” as:

any systematic activity carried on by co-operation between an employer and worker (whether such worker is employed by such employer directly or by or through any agency, including a contractor) for the production, supply or distribution of goods or services with a view to satisfy human wants or wishes (not being wants or wishes which are merely spiritual or religious in nature), whether or not — (i) any capital has been invested for the purpose of carrying on such activity; or (ii) such activity is carried on with a motive to make any gain or profit.

The definition then expressly excludes:

• Institutions owned or managed by organisations wholly or substantially engaged in any charitable, social, or philanthropic service.
• Any activity of the appropriate Government relatable to the sovereign functions of the government, including defence research, atomic energy, and space.
• Any domestic service.
• Any other activity as may be notified by the Central Government.⁹

The Code’s definition is substantially derived from the unnotified 1982 amendment and closely mirrors the Bangalore Water Supply triple test in its positive formulation. The three elements — systematic activity, cooperation between employer and worker, production or distribution of goods or services — are directly reflected in the statutory language. The profit-motive irrelevance holding from Bangalore Water Supply is also expressly codified.

The significant difference is the set of express exclusions. Under the old Act, the exclusions were judicial constructions derived from Bangalore Water Supply. Under the Code, they are statutory. The charitable/philanthropic exclusion is particularly important: an institution “wholly or substantially engaged” in such service falls outside the definition. Whether a particular entity meets that description requires a factual inquiry into the dominant character of its activities — the dominant-nature test from Bangalore Water Supply continues to operate, now as a framework for applying the statutory exclusion rather than for construing a broad inclusive definition.

Practical implications of the Code definition

For purely sovereign government functions — defence establishments, atomic energy departments, space research organisations — the exclusion is unambiguous. For charitable hospitals, the analysis depends on whether the institution is wholly or substantially engaged in philanthropic service or whether it operates as a commercial enterprise with a philanthropic label. A small NGO-run clinic is more likely to fall within the exclusion than a large corporate-backed hospital trust that prices its services at market rates.

For educational institutions, the analysis similarly requires an assessment of dominant character. A small school run by a religious trust for community service has stronger grounds for exclusion than a large private university charging fees competitive with corporate entities.

For gig platforms, technology companies, and emerging service enterprises, the triple test still requires assessment. An organised platform that coordinates the work of thousands of workers through digital infrastructure satisfies the systematic activity and cooperation elements. Whether those arrangements constitute an employment relationship — which is a distinct question governed by the definition of “workman” — affects whether the entity falls within the Act. But the platform itself may well be an “industry” even before that question is resolved.⁹

VIII. Practical Guidance — Applying the Law Today

For disputes arising before 21 November 2025 under the IDA: the triple test from Bangalore Water Supply is the operative authority. Apply the three elements, assess the dominant nature of the enterprise for mixed-function entities, and check whether the specific entity falls within the narrow exclusions the Court recognised — sovereign functions, individual practitioners, or domestic service.

For disputes arising under the Industrial Relations Code: begin with the statutory text of Section 2(p). Map the facts against the express definition and the express exclusions. Use Bangalore Water Supply and the cases that followed it as interpretive aids — the nine-judge bench’s forthcoming judgment will clarify how much of that case law survives in the Code’s framework.

For practitioners conducting due diligence — on acquisitions, restructurings, or workforce planning: the most consequential question is usually whether the entity triggers Chapter V-B obligations (prior government permission for lay-off, retrenchment, and closure). Under the Code, the threshold is three hundred workers. If the entity is an “industry” with three hundred or more workers, every significant workforce change requires government permission. If it is not an “industry,” or if the workforce is below the threshold, ordinary contract law and service rules govern. That distinction defines the entire compliance architecture of large-scale employment.

The nine-judge judgment is the most significant forthcoming development in Indian labour law. If it narrows Bangalore Water Supply, a large number of entities that have operated as “industries” for decades may find that the statutory protections available to their workers change overnight. If it upholds the broad view, the Code’s express exclusions become the primary battleground. Either outcome will require practitioners to reassess their existing advice to clients in the affected sectors.

IX. Key Cases

D.N. Banerji v. P.R. Mukherjee AIR 1953 SC 58 — first major departure from commercial-character requirement; municipality as industry.³

State of Bombay v. Hospital Mazdoor Sabha AIR 1960 SC 610 — hospital as industry; first articulation of dominant-nature analysis.⁴

Bangalore Water Supply and Sewage Board v. A. Rajappa (1978) 2 SCC 213 — seven-judge Constitution Bench; triple test; dominant-nature test; profit motive irrelevant; binding authority for IDA-era disputes.⁵

CESC Ltd. v. Subhash Chandra Bose (1992) 1 SCC 441 — application and confirmation of triple test post-Rajappa.

State of U.P. v. Jai Bir Singh — reference to nine-judge bench; Bangalore Water Supply doubted; judgment reserved 19 March 2026.⁷

X. Exam Architecture

20-mark question: “Critically examine the definition of ‘industry’ under the Industrial Disputes Act, 1947 with reference to decided cases.”

Para 1 (~50 words): The definitional question as the threshold issue for the entire Act. Section 2(j) as the gatekeeper — who is inside the Act’s protective architecture.

Para 2 (~80 words): The two-limb structure of Section 2(j). “Undertaking” and “service” as expansive terms. Pre-Bangalore Water Supply uncertainty — three phases of doctrine (commercial test, D.N. Banerji, functional approach).

Para 3 (~200 words): Bangalore Water Supply — the triple test stated precisely. Systematic activity. Cooperation between employer and employee. Production or distribution of goods or rendering of services. Dominant-nature test for mixed enterprises. What was included. What was excluded (sovereign functions, individual practitioners, domestic service). Profit motive irrelevant — the most significant holding. Apply to illustrative examples.

Para 4 (~100 words): The normative case for the broad interpretation — Kahn-Freund and the structural inequality argument. Workers in non-commercial enterprises face identical power asymmetry. Legal form should not override substantive reality.

Para 5 (~100 words): The criticism — Jai Bir Singh reference. The legislative history corrected: the 2010 amendment did not touch Section 2(j). The unnotified 1982 amendment — what it would have done. The nine-judge bench and its reserved judgment.

Para 6 (~80 words): The Industrial Relations Code, 2020 — Section 2(p). Codification of the triple test. Express exclusions for charitable/philanthropic institutions, sovereign functions, domestic service. The dominant-nature test as the framework for applying the exclusions. Continuity of the doctrinal framework under new statutory language.

The author is a law student at Law Centre-1, Faculty of Law, University of Delhi. Views are personal.