Four Codes, Thirty-Six Rulebooks — The Labour Reform That Is Not Yet a Reform

Pritiraj BrahmaFounding Partner, Pritiraj & Partners · 4 June 2026

Twenty-nine central labour laws were repealed on 21 November 2025. The central rules arrived on 8 May 2026. The states are where the reform is now decided.

ProvenancePublished 4 June 2026; revised 31 August 2026 to add two figures and a provenance line. State rule-making status is as at May 2026 and changes continuously — the six states named are those recorded as having notified final rules at that date, and readers should verify current status before relying on it. The 3 to 15 per cent statutory cost range is a practitioner estimate, not an official projection; no government impact assessment of the wage-definition change has been published. The four codes are the Code on Wages, 2019, the Industrial Relations Code, 2020, the Code on Social Security, 2020 and the Occupational Safety, Health and Working Conditions Code, 2020. Data vintage note: because rule-making is live, the asymmetry this essay describes is a moving position rather than a settled one.

The promise of the labour codes was one rulebook instead of twenty-nine. What employers have today is one code and up to thirty-six sets of rules under it. Simplification is not a statute; it is a state of the world.

On 21 November 2025 the Ministry of Labour and Employment brought all four labour codes into force — the Code on Wages, 2019, the Industrial Relations Code, 2020, the Code on Social Security, 2020, and the Occupational Safety, Health and Working Conditions Code, 2020 — repealing twenty-nine central labour laws in a single stroke. It was the largest consolidation of Indian labour law since independence.

Draft central rules were gazetted on 30 December 2025 for stakeholder comment, with windows of thirty days for the Industrial Relations Code and forty-five days for the other three. The final central rules were notified on 8 and 9 May 2026 — roughly five and a half months after the codes themselves took effect.

That gap is the story, and it is not over. Labour is a Concurrent List subject. Every state and Union territory must frame and notify its own rules. Until it does, employers in that jurisdiction operate under the codes read together with the central rules and whatever legacy state rules survive by virtue of the repeal-and-savings provisions.

Four findings anchor this analysis:

1. The codes are in force; the misconception that they are not is doing real damage. A widely held view in the market has been that the codes await full rule-making before they bite. They do not. A substantial set of provisions operate either independently of rules or on the basis of the surviving rules of the subsumed legislation, and have done since day one. Establishments that paused compliance pending "final clarity" have been non-compliant for six months without knowing it.
2. State notification is asymmetric, and asymmetry is the cost. By May 2026, several states — among them Gujarat, Haryana, Madhya Pradesh, Karnataka, Maharashtra and Arunachal Pradesh — had notified final state rules, while most others remained at draft or pre-publication stage. A multi-state employer therefore maintains a state-keyed compliance matrix: the same workforce policy is compliant in one state and defective in the next.
3. The wage-definition change is the operative economic provision. The requirement that basic wage constitute at least half of total remuneration restructures provident fund, gratuity and insurance liabilities across the formal sector. Practitioner estimates put the increase in statutory cost at roughly 3 to 15 per cent depending on the existing salary structure. For labour-intensive sectors operating on single-digit margins, that is a strategic variable, not a payroll adjustment.
4. Fixed-term employment has been quietly repriced. Gratuity eligibility for fixed-term employees at one year rather than five changes the cost calculus of project-based hiring — the arrangement on which construction, infrastructure delivery and much of India's contract manufacturing depend.

Why the transition is harder than the drafting was

Consolidating twenty-nine statutes is an act of legislative craft. Landing them in thirty-six jurisdictions is an act of administrative capacity, and the two are not the same skill.

Three mechanics make the transition harder than it looks on paper. First, the surviving-rules doctrine: where old rules conflict with the codes, the codes prevail — which requires every employer to conduct a provision-by-provision reconciliation between a new statute and a rulebook written for a repealed one. Second, institutional continuity: the first reported High Court judgment under the codes, in the Kerala High Court, addressed tribunal continuity — an early signal that the litigation of this transition will be about which forum and which procedure survives, not about the policy. Third, guidance stability: practitioners have noted that certain clarifications and FAQs issued after November 2025 subsequently ceased to be available on official websites. Compliance guidance that moves is worse than compliance guidance that is late, because it cannot be relied upon at the moment of decision.

The distributional question

A note on what the codes are for. They were enacted to formalise employment, simplify compliance through digital integration, strengthen social security — including for gig and platform workers — and improve workplace safety. Those are worthwhile ends and this Review does not dispute them.

But the sequencing has a distributional consequence worth naming. Large employers have counsel, payroll systems and the capacity to run a thirty-six-jurisdiction matrix. Small and medium establishments — which employ the overwhelming majority of India's formal workforce — do not. In a transition where the binding constraint is interpretive capacity rather than statutory obligation, the burden falls on precisely the firms least able to carry it, and the predictable response is not compliance but informality. A reform intended to formalise employment can, through slow and uneven rule-making, push employment the other way. That risk is manageable, and it is managed at the state level or not at all.

The counter-case, honestly stated

Three objections deserve a hearing.

First, asymmetry is the constitutional design rather than a failure of it. Labour sits on the Concurrent List because wage levels, industrial composition and working conditions genuinely differ between Gujarat and Arunachal Pradesh. A state that adopts the central rules unaltered is not simplifying; it is declining to exercise a power it was given for a reason. The call to notify convergently asks states to treat divergence as a cost requiring justification, which is a defensible policy preference — but it is a preference about federalism, argued here as though it were a neutral point about efficiency.

Second, five and a half months is fast by the standard of comparable Indian transitions, and the criticism may simply be early. The Goods and Services Tax required years of clarifications and its rate structure was still being rebuilt in September 2025, eight years after commencement, as this Review has recorded elsewhere. Against that benchmark, central rules notified within six months and a majority of the large industrial states moving within a further quarter is quick work. A transition still under way is not a transition that has failed, and judging rule-making at month six risks mistaking sequence for defect.

Third, the distributional argument runs both ways, and only one direction is counted. The requirement that basic wage constitute at least half of remuneration raises provident fund, gratuity and insurance entitlements for the workers those small firms employ, and gratuity eligibility for fixed-term employees at one year rather than five transfers real value to project-based workers who previously accrued none. The same provision that imposes a 3 to 15 per cent statutory cost on a marginal employer raises deferred wages for a marginal worker. Whether the net effect is formalisation or a retreat into informality is an empirical question, and the assumption that the predictable response is informality is a forecast presented with the confidence of a finding.

Two recommendations survive all three objections untouched: the single reconciliation document and immutable, versioned guidance. Both are cheap, both are pure delivery, and neither depends on winning the argument about convergence.

What we would do

For a state labour department, the next ninety days determine whether this reform lands as simplification or as friction.

1. Notify, and notify convergently. The value of the codes is national consistency. A state that notifies rules materially divergent from the central rules recaptures the complexity the codes were passed to remove. Divergence should be reserved for genuine local necessity and defended explicitly where used.
2. Publish a single reconciliation document. One authoritative table, per code, showing which legacy state rules survive, which are displaced, and from what date. Every large employer in the state is currently building this document privately and inconsistently. It costs a department one competent team a fortnight; it saves the state's employers thousands of hours and a great deal of avoidable litigation.
3. Put the small-establishment segment on a supported transition. A helpline, a model wage structure worked example, and a stated enforcement posture for the first two compliance cycles. Enforcement that begins with a penalty rather than a notice will convert marginal firms into informal ones.
4. Make guidance immutable. Version every circular, date it, and never withdraw one silently. Employers plan against published guidance; guidance that disappears is guidance that was never usable.
5. Instrument the wage-cost effect before it becomes a political question. The 50 per cent wage rule will raise statutory costs unevenly across sectors within every state. A department that measures that effect in its own labour-intensive industries will be able to shape the national conversation with evidence. A department that does not will receive the conversation as a grievance.

The hard legislative work is done. What remains is the part India has historically found harder: making the same rule mean the same thing in every district that has to apply it.