India's Four Labour Codes: What Actually Changes for Workers

For most of India’s working life, our labour law was a thicket. Twenty-nine central Acts, some dating back to 1923, each with its own definitions, its own registers, its own inspector. If you were a factory hand you were covered by one set of rules; a shop worker by another; a construction labourer by a third; and the vast majority — the street vendor, the domestic worker, the delivery rider — were covered by almost nothing at all. The new labour codes fold all of that into four consolidated statutes, and after years of delay they are finally rolling out. The government notified the codes as law in late 2025 and the central rules through 2026, though actual enforcement still depends on each state finishing its own rules — so this is a staged, uneven rollout rather than a single switch flipped on one clean date. The question worth asking is not whether the codes are tidier. They obviously are. The question is the one that matters for anyone who works for a living: who actually gains, and who quietly loses?

What the four codes are

The 29 old laws collapse into four codes, each covering one domain of working life:

• The Code on Wages — a single law governing minimum wages, payment of wages, bonus and equal remuneration, and, for the first time, a national floor wage below which no state’s minimum may fall.
• The Industrial Relations Code — the rules on trade unions, standing orders, retrenchment, layoffs and strikes; the code that governs the balance of power between employer and worker.
• The Code on Social Security — provident fund, state insurance, gratuity, maternity benefit and, notably, a framework meant to extend social security to gig, platform and unorganised workers.
• The Occupational Safety, Health and Working Conditions Code — safety standards, welfare provisions, working hours, leave and the terms on which contract and migrant labour may be engaged.

Between them, the codes are meant to bring hundreds of millions of workers — the overwhelming majority of the workforce, most of whom were previously outside the reach of formal labour law — under one framework. That reach is real, and it is the strongest thing you can say in the codes’ favour. But reach on paper is not the same as protection in practice, and the gap between the two is where the whole story lives.

The genuine gains

Start with what is honestly better, because there is a temptation on the left to dismiss the whole exercise, and that would be a mistake.

The statutory floor wage is a real structural change. Under the old regime, minimum wages were a patchwork — thousands of separate rates across states, occupations and skill levels, many of them absurdly low and rarely revised. A national floor wage sets a legal ground beneath the entire system: no state minimum is supposed to drop below it. Whether the floor is set at a dignified level is a political fight that will play out every time it is revised, and there is every reason to watch it closely. But the principle — that there is a number below which paying a worker is illegal anywhere in the country — is a genuine advance over the chaos it replaces.

The widening of social security on paper is the second gain. The Code on Social Security explicitly names gig workers and platform workers as categories deserving of cover, and it envisages schemes funded partly by the aggregator companies themselves. For the first time in Indian labour history, the rider ferrying your dinner and the driver taking you home are written into the statute book as workers owed something, rather than as invisible “partners” in an app’s terms of service. I have written before about the specific fight over gig worker social security, and the codes are, at minimum, an acknowledgement that the problem exists at the level of national law.

Reach on paper is not the same as protection in practice, and the gap between the two is where the whole story lives.

Third, there is the plain value of consolidation itself. A worker who wants to know their rights should not need a lawyer to navigate four overlapping definitions of “wages” and three of “employee.” Simpler law is, in theory, more enforceable law — and enforcement, not the elegance of the statute, has always been where Indian labour protection actually failed. If the codes make it easier for a worker to know what they are owed and to claim it, that is worth something real.

The losses critics flag

Now the other side of the ledger, because it is substantial and the trade unions who have staged nationwide protests against the codes are not inventing their grievances.

The most consequential change is the raised threshold for hire-and-fire. Under the old law, a firm employing 100 or more workers needed prior government permission to lay off staff or close a unit. The Industrial Relations Code lifts that threshold to 300. In plain terms: a large swathe of medium-sized firms can now retrench workers or shut plants without asking anyone first. The government frames this as flexibility that will encourage hiring. Unions read it, not unreasonably, as legalised hire-and-fire for a huge new band of employers. Which reading is right depends on empirical outcomes we do not yet have — but the burden of the change falls, unmistakably, on the side of the employer.

Alongside it sits the normalisation of fixed-term employment. The codes give statutory blessing to hiring workers on short contracts — six months, eleven months — with no requirement that the underlying work be temporary and no obligation to renew. A fixed-term worker is entitled to the same wages and benefits as a permanent one for the duration, which sounds fair; but a rolling series of expiring contracts is a structural substitute for the security of a permanent job. It is a way to keep a permanent function staffed by permanently disposable people.

Then there are the tighter conditions on strikes. The codes extend the notice and cooling-off requirements that previously applied only to public utilities to workplaces generally. Workers must give advance notice, cannot strike during conciliation proceedings, and must wait out a further period afterward — and stepping outside these windows can expose them to penalties. A strike that must be announced weeks ahead and cannot be called during the very negotiations it is meant to pressure is a strike with much of its leverage removed. The right to withdraw labour is the ultimate source of a worker’s bargaining power; hedge it enough and collective bargaining becomes collective petitioning. This is exactly the terrain I explored in writing about gig workers unionising — the capacity to organise and to act collectively is the thing that turns a paper right into a real one, and it is precisely that capacity the strike provisions constrain.

Finally, the codes permit longer working hours. The rules allow daily shifts to be lengthened — the widely discussed twelve-hour working day — with the promise that weekly hours and overtime rules still apply and that longer days come with more days off. In a labour market with weak enforcement, “flexibility” over daily hours has a way of becoming a longer day at the same weekly pay. Whether the safeguards hold will, once again, depend entirely on inspection and enforcement that has historically been thin.

The pattern underneath

Step back and a shape emerges. The gains in these codes — the floor wage, the social-security cover — are provisions that require the state to build machinery, fund schemes and enforce standards, and their real value will only be known years from now, when we see whether the machinery was actually built. The losses — the higher retrenchment threshold, the fixed-term contracts, the strike constraints — are effective the moment they are notified, because they consist of removing restraints on employers rather than creating obligations. Deregulation lands immediately; protection is deferred to implementation. That asymmetry is not an accident of drafting. It is the structural bias of the whole reform.

Deregulation lands immediately; protection is deferred to implementation. That asymmetry is not an accident of drafting.

This is a version of a pattern I keep returning to in thinking about how technology gets captured: a reform is sold on the language of the powerless — here, the fifty-crore uncovered workers who will finally be brought inside the law — while its operative provisions transfer power to the already-powerful. The gig worker written into the statute is real. So is the aggregator’s freedom to keep classifying that same worker in ways that minimise what it owes. Both live in the same codes. The framing foregrounds the first and lets the second do its quiet work.

None of which means the codes are simply a fraud on workers. That would be too easy, and it would ignore the floor wage and the social-security architecture, which a genuinely cynical government would not have bothered to write at all. The more honest reading is that the codes are a bargain: real, if uncertain, gains in coverage traded for real, immediate losses in security and bargaining power. Whether that bargain favours workers depends almost entirely on things the codes themselves do not guarantee — the level at which the floor wage is set and revised, whether the gig-worker schemes are actually funded and paid out, and whether labour inspection is resourced enough to make any of the protections bite.

What decides the outcome

Here is the part worth holding onto. Laws of this kind are not self-executing, and their meaning is settled after they pass, not before. The floor wage is a number that gets fixed in a room; who is in that room matters. The gig-worker welfare fund is a scheme that gets designed and financed; whether workers have a voice in it matters. The strike rules are constraints that get tested in practice; whether unions retain the organisational strength to push back matters. The codes set the field. They do not decide the game.

That is why the outcome will turn less on the text than on the balance of forces around it — on whether working people remain organised and mobilised enough to make the good provisions real and to blunt the bad ones. India has seen what sustained, disciplined collective pressure can do to a policy the government considered settled; the story of how India’s farmers won the repeal of the farm laws is the clearest recent proof that a determined movement can move an immovable state. The labour codes are now law, and repeal is not the likely path. But the terms on which they operate — the numbers, the schemes, the enforcement — are all still open, and they will be decided by who shows up to contest them.

So when someone tells you the new codes are either a historic gift to the Indian worker or a straightforward assault on their rights, be suspicious of both. They are a framework with real gains and real losses baked into the same pages, and the direction they finally tilt has not yet been determined. It will be determined by us — by whether the people the codes name as beneficiaries have the collective strength to make the naming mean something. That was always the real question. The four codes just make it unavoidable.