One rule, thirty-six variants: why Indian compliance does not scale

A manufacturer opens a second plant. Same product, same process, same standard operating procedures, same HR policy. The compliance team assumes the obligation set is a copy of the first plant's.

It is not, and the gap is larger than anyone budgets for.

What actually differs

The second plant is in a different state. That single fact changes:

• The Shops and Establishments or Factories Act registration, its form, its fee and its renewal cycle
• The registers to be maintained and their prescribed formats
• The annual return, its due date and in some states whether a half-yearly return also applies
• Notified minimum wages, by skill category, revised on the state's own schedule
• Professional tax slabs, and whether professional tax applies at all
• Labour welfare fund contribution, rate and periodicity
• The Pollution Control Board, its consent procedure, its validity period and its conditions
• Municipal trade licence, fire NOC and, in some corporations, a separate factory trade licence
• The inspector who will visit, and the interpretations that inspector applies

None of this is exotic. All of it is routine, and all of it is different.

Why consolidation did not fix this

Both of the big simplification reforms of the last decade are genuinely simplifications, and both operate at a level above where compliance actually happens.

GST unified the tax. It did not unify the compliance. A company registered in fourteen states files fourteen sets of returns, faces fourteen jurisdictional officers, and receives notices from fourteen directions. The tax is one tax; the compliance is fourteen compliances.

The Labour Codes consolidated twenty-nine Acts into four. The four are easier to read. But registration, registers, returns, thresholds and inspection remain with the states, and the state rules notified so far diverge on all five. An employer in five states still runs five regimes; they are just described by four statutes instead of twenty-nine.

This is not a criticism of either reform. It is a structural feature of a federal system in which labour and several other subjects sit on the Concurrent List and much administration sits with the states. It is not going to change, and compliance operating models should stop assuming it will.

The three mistakes this produces

Copying the compliance calendar

The most common. Plant two inherits plant one's calendar, and the differences surface at the first inspection. Typically the missing items are the state-specific ones: a register nobody knew was prescribed, a half-yearly return the first state does not require.

Centralising the wrong thing

Some organisations respond by centralising all compliance into a corporate team. That team then cannot answer state-specific questions and becomes a bottleneck without becoming a control.

What should be centralised is the register — the definitive statement of what applies where. What should stay local is execution, because the person who deals with the state inspector needs to own the obligation.

Assuming Central law is the whole obligation

Most compliance registers we are asked to reconcile are built from Central legislation, because Central legislation is what is easy to find and read. The state layer is between forty and sixty percent of the actual obligation count for a multi-site organisation, and it is the layer most often missing entirely.

What works instead

Model the site, not the company. The unit of scoping is a site in a state, with its own headcount, category, process and licences. A company-level register cannot express the differences that matter.

Treat the state delta as first-class. For each Central obligation, hold the state variant explicitly — what differs, in which state, and why. Do not flatten it into a lowest-common-denominator obligation, and do not maintain thirty-six unrelated registers either.

Assign locally, govern centrally. The plant HR manager owns the state return. The corporate team owns knowing that the return exists, that it has an owner, and whether it was filed.

Expect the register to be the deliverable. For most organisations doing this seriously for the first time, the single most valuable output in year one is simply an accurate list of what applies at each site. Everything downstream depends on it, and almost nobody has it.

The Indian compliance problem is not that the rules are unusually harsh. It is that they are unusually numerous, unusually distributed, and unusually badly documented in one place. That is a data problem before it is a legal one.