Traffic challan rules in every Indian state and union territory: one question, thirty-five answers, counted.

What this page is

Every other traffic page on this site is about one place. This one is about the comparison, which is the thing none of them can do on its own.

Over thirty-five pages we read one government’s instruments at a time — its notifications, its rules, its transport department’s own web pages, its police force’s own web pages — and asked a single question of each. This page puts all thirty-five answers next to each other and then says what the arrangement looks like from above: how many governments have done the thing the Act asks of them, how many are working from documents older than the law they price, where the enormous differences are, and which half of the job almost nobody does.

It also carries something no individual page can carry properly: a consolidated record of what we got wrong across the series, what we did about each, and what is still outstanding. That is at #ourowncorrections and we would rather you read it than find it.

The one question, put thirty-five times

Here it is, in the form we used every time, so you can see it is the same question and not a moving target.

We chose that question because it is the only one a reader can act on. You cannot audit a state’s road safety policy from a kerbside. You can check a number against a published document, if there is one.

Why section 200 is the whole question

Almost every interaction an ordinary person has with Indian traffic law happens under one provision, and most people have never heard its number.

Section 200 of the Motor Vehicles Act, 1988 permits certain offences to be compounded — settled by a payment to a specified officer, which closes the matter without a magistrate. That is what is happening when you pay at a counter, on a portal, or to an officer by the roadside. Without section 200, every helmet offence in India would be a criminal case in a court, which would be unworkable for everybody including the courts.

So compounding is a good thing, and the Act asks something in return for it.

The two halves nobody separates

Section 200 requires a government to notify two things, and this distinction runs through every page on this site.

One more thing about section 200, which took us twenty-one pages to establish by reading the provision itself rather than commentary: the compoundable list is Parliament’s, not any state’s. A state cannot add an offence to it or take one off. That is why drink driving under section 185 can never be settled at a counter in any part of India, however the local arrangement is described to you. We set that out in full on the Meghalaya page, which is where we finally read it.

All thirty-five, in one table

Every row links to that jurisdiction’s own page, where the instrument is set out and quoted. The dates are the dates of the instruments, not of our reading. Where a column says we could not find something, that is what it means — not that it does not exist.

How to read that table

Three things about it, because a table like this can mislead if you read it too quickly.

And the third thing: the dates are the dates of the instruments. That column is the one worth staring at, because the Motor Vehicles (Amendment) Act, 2019 commenced on 1 September 2019, and everything above that line in that column is a document pricing penalties that have since been rewritten.

The scorecard

Four numbers, counted off the table above rather than estimated.

We have written the third one with a tilde because the boundary is a judgement call: a settlement scheme or a set of paperwork composition rates is something, and a reader deciding whether their jurisdiction “publishes” could reasonably count one or two of the amber entries either way. The two solid numbers are five and seven, and the one that genuinely shocked us is two.

The five that published after 2019

These are the only jurisdictions on this site whose compounding instrument postdates the amendment that rewrote the penalties it prices.

Four of the five are within about two years of the amendment, and three of those within four months of it. Which tells you that where a government decided to do this, it did it quickly — the drafting is not the obstacle.

The seven running on documents older than the Act

And here is the other side of it. These jurisdictions have a schedule; it prices offences whose penalties Parliament rewrote after the schedule was written.

The Mizoram entry is the sharpest illustration of the problem, because it is only a year older than the amendment and it still lists a section that no longer exists while listing nothing that 2019 inserted. A document does not have to be ancient to be wrong; it only has to be on the wrong side of one line.

Why staleness matters more than lowness

It would be easy to read the stale list as “these states charge too little”. That is not the problem and it is worth being clear about why.

This is the strongest practical argument for the drafting technique we recommend throughout this site: a notification that adopts the Act’s compoundable list and figures by reference instead of transcribing them cannot go stale, because when the Act moves it moves.

The twenty-odd that publish nothing

Around twenty-one of the thirty-five publish no compounding amount for road offences that we could find. Rather than list them again — they are in the table, in red — it is more useful to say what their pages turned out to be about, because in almost every case the absence was not the most interesting thing we found.

That is the honest shape of this series. We went looking for schedules, found far fewer than we expected, and the pages ended up being about whatever each government had done instead. In every one of those jurisdictions the only route to a figure is still asking for the notification in writing, which is why that is the instrument we end up drafting most often.

Not all silence is the same

We have tried not to write the same angry paragraph twenty times, because the twenty cases are not alike and pretending they are would be lazy.

And two cases sit oddly in between. One is a territory where two unpublished rulebooks are alive at once, so a person cannot even identify which silence applies to them — we explained why we think that is worse than a single silence on the Dadra & Nagar Haveli and Daman & Diu page.

The other is Punjab, and it is worse again in a way we did not anticipate when we started counting. Punjab’s silence does not produce thin results; it produces confident wrong ones. Because there is a Punjab province in Pakistan running a well-indexed camera-based e-challan system out of Lahore, a search for “Punjab e-challan” is answered by a foreign authority and a row of commercial sites, with no Indian government result at all. A reader who finds nothing at least keeps an accurate picture of the world. A reader who enters a vehicle number into the wrong country’s system and is told there is no challan does not.

The officer half, state by state

This is the half of section 200 that gets forgotten, and it produced the widest variation of anything in the series — wider than the money.

Ten jurisdictions out of thirty-five publish something about the officer half. Two of them — Haryana and Rajasthan — publish the officers and not the money, which is the mirror image of the usual failure and at least as odd.

From a Police Constable to one man in a state

Put the two ends of that table next to each other, because the distance is remarkable.

Neither is wrong. A low floor makes roadside settlement practical and a high floor makes a serious offence hard to settle casually, which is probably the point in Bengal’s case. What the contrast shows is how much room section 200 gives a government — and therefore how much a reader loses when the government does not say which choice it made.

Why the rank matters to you

Not as a technicality. As the thing that determines whether the payment you made did what you think it did.

Two schedules in India mention your licence

Of everything in this series, this is the finding we were least prepared for.

Two provisions cost a driver three months of their licence as well as the money: section 194D, for a bare head, and section 194C, for one person too many on a two-wheeler. Nothing a traffic officer deals with is more routine than those two, and they are settled at counters across India every day for a thousand or two thousand rupees.

Two mentions, in thirty-five jurisdictions, of a consequence that attaches to the two offences people are stopped for most often. Andaman & Nicobar comes closest to a third, noting on relevant rows where the licence is to be seized, which is a different thing but the same instinct.

Why we think that is the biggest gap of all

Bigger than the missing amounts, and we want to explain the reasoning rather than just assert it.

And it is cheap to fix. Chandigarh’s solution is one extra note on the relevant rows of a schedule it was drafting anyway. That is the whole intervention.

What a schedule is actually for

We started this series assuming the answer was obvious and we were wrong about it. The correction came on the Tripura page and it is set out at #thetripurareversal below. What came out of thirty-five pages is that a published schedule does four distinct jobs, and most arguments about traffic fines confuse them.

One: the citizen’s protection

The first and most important job, and the one Tripura taught us. A published schedule is the only thing that lets a person check a demand for money against a figure somebody else wrote down.

Without it, the only check available is the central Act’s ceiling, which is a blunt instrument: it will catch an officer asking for twenty thousand rupees for a helmet, and it will not catch one asking for a thousand where the state’s own policy is three hundred.

And it converts a grievance into a claim. You cannot substantiate an allegation of overcharging if nobody will tell you what the correct charge was. That is the whole of the argument and it does not need any road-safety framing to stand up.

Two: traceability

The second job came out of Sikkim, which has the oldest price list in the series and, simultaneously, the most citizen-protective automated enforcement arrangement we found anywhere: a ten-day window to cure, SMS and WhatsApp alerts, and a written undertaking to compensate you with double the penalty if the machine got it wrong.

Put those two facts together and the point emerges. A schedule tells you which rulebook produced your figure. In a world of automated challans generated by software, that is not academic: it is the difference between a number you can trace to an instrument with a date on it and a number that simply appeared.

Three: routing — counter or court

The third job is the one almost nobody designs for, and two jurisdictions showed us why it matters, from opposite directions.

Maharashtra does the same job from the other end, by publishing a list of nine offences that cannot be compounded at all — the only state that published what you cannot pay for. Two of those are things neighbouring states let you settle for money.

Four: an input to what the money is spent on

The fourth job is the one that turns a compounding schedule from an enforcement document into a budget line, and two jurisdictions have done it.

Ladakh’s is the strangest and most interesting document in the series, because it directs a share of the money without the amounts ever having been published. The same notification defines a compounding fee as one taken by “the officers authorised under Section 200” — so that authorisation exists somewhere, and we still could not find it. A territory has published where a fifth of your money goes, and not what the money is.

Patterns we did not expect

Two patterns emerged that we were not looking for, and neither is the one you would guess. Both are about which governments publish nothing, and both cut against the obvious assumption that it is simply a matter of size or capacity.

The tax pattern, five times over

Five small jurisdictions turn out to have exactly one motor vehicles enactment of their own, and in every case its subject is taxation.

We do not think this is a conspiracy, and we want to say what we do think, because the honest explanation is duller and more useful than a sinister one.

The union territories, the wrong way round

The second pattern is the one that genuinely surprised us, and it came out of the last five pages in the series.

Section 200 gives the power to “the State Government”. In a union territory that phrase is complicated: the General Clauses Act says it means the Central Government, while a Council of Ministers advises an Administrator. We set the provisions out on the Puducherry page and left the question open; Chandigarh then answered it in practice, by issuing a notification that simply states the power is exercised by the Administrator — who since 1984 is, ex officio, the Governor of Punjab.

You would expect the territories with their own legislatures to be the ones that managed it. The opposite is true.

The two with legislatures published nothing. Of the five without, two published something substantial and one published a funding rule. We are not going to build a theory on seven data points, but it does dispose of the idea that the problem is a missing law-making power, because the places with the most law-making power are the places where nothing was made.

The ceiling, the floor and the formula

Three jurisdictions did something with the structure of the amounts rather than the amounts themselves, and together they describe the whole space a drafter is working in.

Odisha’s is the one we find most impressive as an act of drafting discipline, because somebody went through seventeen rows distinguishing a discretionary maximum from a mandatory minimum and got it right in every single one. That is careful work and almost nobody would ever notice it.

The one that aged well, and why it still failed

Meghalaya’s formula deserves its own section, because it is the closest anybody in India has come to the drafting technique we keep recommending — and it shows exactly where that technique stops helping.

The hundred rupees that is Parliament’s, not the states’

A figure that recurs across several states’ rules and that we misread three times before Goa put us right. The correction is listed below with the others; here is what the figure actually is, because it comes up constantly.

Several states compose a paperwork lapse — a late registration renewal, a transfer not reported — at a sum “not exceeding one hundred rupees”. Karnataka charges twenty-five rupees a calendar month to that cap. Uttarakhand counts weeks and prices by vehicle, so a lorry hits the ceiling in a fortnight and an invalid carriage takes two years. Tamil Nadu uses bands of days.

Ten to one, for the same words

Where amounts are published, the spread for identical conduct under identical statutory words is larger than most people would guess.

The Maharashtra row is the interesting one, because the cheapest option on that row is not cheap at all. An offence with no compounding amount is not a bargain; it is a court date, both the driver and the owner charged, and a minimum three-month sentence on a repeat. “No price” is the most expensive entry in that column.

What that gap does and does not prove

The change in January 2025 that nobody announced

Twenty-four pages into this series we found something that affects every schedule in the country, and exactly one government had told the public about it.

Following section 177A back led somewhere stranger still. The driving regulations it punishes were superseded in 2017, and state departments are still publishing the 1989 version. So a provision that became compoundable in 2025 punishes breaches of regulations that several governments are circulating in an out-of-date form.

What that means for every schedule in India

Which is the final, and we think decisive, argument for the technique we keep recommending. A notification that adopts Parliament’s compoundable list by reference would have picked those three up automatically on 13 January 2025 without anybody lifting a pen. Every schedule that transcribed the list instead now needs amending, and in sixteen months none has been.

There is also a smaller casualty in that substitution, which we flagged on the Nagaland page: the community service power that the earlier text contained may have been deleted before any state in India ever used it. We could find no instance of its use. That is a finding about an absence and we hold it loosely.

The three numbers governments almost never publish

Separately from the schedules, there are three numbers a person might reasonably expect to be able to find, and almost nobody publishes any of them.

Two further figures are worth recording because they describe the direction of travel. Uttarakhand reported a year in which challans rose 118.77% while collections rose 37.07% — issuance running at roughly three times the rate of recovery. And a Supreme Court order of 28 September 2026, which we wrote up on the Jharkhand page, turns an unpaid challan into a blocked registration, which changes the recovery picture considerably.

Our own corrections, in one place

Thirty-four pages is enough work to get things wrong in, and we have. Rather than leave the corrections scattered across the pages where they happened, they are collected here, with what we did about each and what is still outstanding.

We publish these for a simple reason. This whole site asks you to check government documents rather than take a figure on trust. We cannot ask that and then quietly fix our own mistakes.

The three statutory figures we got wrong

The most serious one, because these appeared in reference tables that readers would have relied on.

How we caught it is the part worth keeping. A police force in Port Blair had it right and we had it wrong. Row 17 of the Andaman & Nicobar spot fines list prices section 190(2) at ten thousand rupees with the correct subject-matter, and rows 4 and 5 price section 192(1) in a graduated band from two to five thousand — the first-offence range exactly, used end to end and never exceeded. When a government document and our table disagree, we go to the Act. We then checked against the amending Act of 2019 itself rather than a commentary, which is a rule we made for ourselves earlier in this series after being misled by one.

A small territorial police force got two provisions of the Act right while our own reference tables had them wrong in four places. There is no flattering way to put that, and we are not going to look for one.

What we did about it, and what is still outstanding

This is the part most sites would leave out, so we are putting it in the main flow of the page.

The “fastest in India” claim

A smaller one, repeated across several pages, which makes it worse rather than better.

The hundred-rupee misreading

Three pages described the hundred-rupee composition ceiling in various states’ rules as a stale state figure that nobody had bothered to update. That was wrong, and Goa is where it came apart: its rule 54 names its own parent provision, section 50(3) of the central Act. The ceiling is Parliament’s.

The substance of the finding actually got stronger once corrected — a thirty-one-year symmetry between section 177 and section 50(3) became a fifteen-to-one gap in 2019 because Parliament raised one and not the other — but we had attributed it to the wrong legislature, and the pages that said so were amended.

The Tripura reversal

The one that changed what this series is about.

Early pages implied, and in places stated, that publishing a compounding schedule was a road safety measure. The Tripura page is where that stopped being tenable: its accident figures showed deaths falling about forty per cent in two years while crashes fell nine. That is a survival story, not a prevention one — better emergency care, better trauma response, something downstream of the collision — and it has nothing to do with whether a schedule is on a website.

The Sikkim finding we withdrew

The knock-on from the three wrong figures, and the most substantial withdrawal in the series.

On the Sikkim page we wrote that two rows of its 2010 schedule sat below the statutory minimum, and built a section around the legal question that raised — whether a state may fix an amount below a floor Parliament set.

A sixth one, found while fixing the others

Correcting the four tables meant reading every line on those pages that touched the three provisions, and that sweep turned up something we had not noticed.

This is the kind of thing that only surfaces when a correction is done properly rather than by swapping two numbers and moving on. It is also the sixth item on a page that started with five.

Why these are on the page and not in a footnote

Because of what this site is asking you to do.

Every page here tells you to find the section, find the instrument, read it yourself and check the number against it. That instruction is worth nothing from a source that is not itself checkable. If we corrected our tables silently, you would have no way of knowing which version you had read or whether anything else had been changed underneath you.

So the rule we keep is: a correction goes where the error was, in the same type size as everything else, and it says what was wrong rather than quietly becoming right. This page is where they are all gathered so that one link shows you the whole record.

What we refuse to use

One rule explains more about these pages than anything else, including why so many of them have no figures on them.

Those sites are the first ten results for almost every traffic query in India and most of them carry a confident table of penalties. The tables are frequently pre-2019, frequently copied from each other, and never cite an instrument. A number with no source cannot be checked, and a page built on unverifiable numbers is worse than a page that admits it does not know.

What that rule costs us

The same rule applies to documents we could not open, and we name those too. India Code failed us five times across this series. Tripura’s Acts link points at a file that will not open and its Notices link points at nothing. Several state gazette archives are unreachable. One industry association reproduced what it called a transport department notification and we did not use it, because it is not a government source. Those are all on the pages where they arose.

What you can check anywhere in India

Enough about governments. If you are holding a piece of paper, three things are checkable wherever you are and whatever your jurisdiction has or has not published.

The Act’s own figures

For most people in India this is the operative table, because their jurisdiction has published nothing narrower. These are our corrected figures — see #thethreefigures for what we had wrong and how we found out.

Two rows of that table are worth reading twice, and they are the two in the middle. Sections 194C and 194D are the cheapest serious offences in the Act in money terms and the most expensive in consequence, because three months of licence attaches to both — and, as set out above, thirty-two of thirty-five jurisdictions never mention it.

These are maximums and not tariffs. An officer may take less, and in the states with published schedules many do. Nobody may lawfully take more, and that holds in every one of the thirty-five jurisdictions on this site.

What can never be compounded, anywhere

This list is Parliament’s and no state can shorten or lengthen it. If one of these is printed on your paper, no counter payment closes the matter in any part of India.

The section 184 carve-out catches people out constantly. Dangerous driving generally has to go before a magistrate, but the limb about using a hand-held communication device while driving can be settled at a counter. Andaman & Nicobar handles this better than anybody, pricing the phone limb at three thousand rupees and marking the other five limbs for court in the same row, with the original licence to be seized.

And note that Maharashtra is the only state that published its own version of this list — nine non-compoundable offences, two of which its neighbours let you settle for money.

Reading your own challan

In order, and it takes about ten minutes wherever you are.

One addition for a single jurisdiction, which we mention because it applies nowhere else: in Dadra & Nagar Haveli and Daman & Diu, write down which district the paper came from, because section 17 of its merger statute means the law that applies to you depends on which half of the territory you were in.

What is worth asking in writing

In the twenty-odd jurisdictions with nothing published, an information request is the only route to an answer. The statutory fee is ten rupees and we draft one for five hundred.

Where a jurisdiction was formed out of others, add a fourth: whether any such notification was issued by the predecessor State or territory and is treated as continuing in force. Without it, a truthful “no” about the new entity can leave you knowing nothing — which is exactly the trap on the Dadra & Nagar Haveli and Daman & Diu page.

If the answer is that no instrument exists, that is worth having on paper. It is a documented instance of a central power that has never been operated in part of India, and it only becomes visible when somebody asks in the right words. For the cases where a reply has to be answered rather than just obtained, there is a drafted reply, and for the plainer asks a simple application.

The picture region by region

Not because geography explains much, but because readers tend to want to know how their part of the country compares, and a few genuine regional shapes did emerge.

Punjab joined this list late, and for a reason that was ours rather than the state’s: an early note recording that its transport website would not load for us hardened into a label that outlived the problem, and nobody re-tested it. It is now the thirty-fifth page, and it turned out to be the only jurisdiction in India where the search results themselves mislead — a query for a Punjab e-challan is answered by the Punjab Safe Cities Authority in Lahore, in Pakistan, because Indian Punjab has published nothing for the search to find. Manipur is still not here, so this page says thirty-five and not thirty-six. We are not going to round the number up.

What would fix the whole thing

One document, repeated about twenty times, and it would fit on two pages each.

Two jurisdictions have already shown the pieces work. Assam set all eighty-eight of its rows at the Act’s own figure, which proves the third row is workable. Chandigarh wrote the disqualification in, which proves the fourth is a matter of one extra note. Meghalaya showed a formula survives an amendment, and its one failure — a frozen list of nine sections — is precisely what row two exists to prevent.

No survey. No consultation. No money. One officer and one afternoon in each jurisdiction that lacks one.

Why it has not been done

We have thought about this across thirty-five pages and we do not believe the explanation is sinister. We think it is duller and harder to fix for exactly that reason.

What this page cannot do

It cannot tell you your local figure, because for most readers of it no local figure has been published. It cannot tell you what an unpublished instrument says. It cannot cover Punjab or Manipur, which are not in this series yet. And it is not a substitute for the individual page: everything here is a summary, and the instrument, the quotation and the detail are on the jurisdiction’s own page, which every row of the table links to.

What it does give you is the ceiling Parliament set, Parliament’s own list of what a counter may settle, the two provisions that take three months of licence along with the money, your jurisdiction’s place among the other thirty-three, and a request worded so that it cannot be answered with a technically true nothing.

What we could not establish

Our standing practice is to name the doors we could not get through, because a page that reports only what it found is a page you cannot calibrate.

How this page was put together

Entirely from the thirty-five pages it indexes, each of which was built from that government’s own instruments and web pages and from the central Act. The table was compiled by going back through them one at a time rather than from memory, and the counts in the scorecard were taken off the finished table rather than estimated.

The list of jurisdictions on this page is generated from the same registry that builds the site’s own index, so it cannot drift out of step with the pages that exist. Where this page summarises a finding, the instrument behind it is quoted on the jurisdiction’s own page and the row links there.

No commercial challan or rate-list website was used for anything, here or on any of the thirty-five.

When a document stops being the answer

Most of this site is about paper, because most challans are a paper problem. Some are not, and it is worth being clear about which.

If the section on your challan cannot be compounded — drink or drugs under 185, leaving a collision under 187, dangerous driving under 184 other than the phone limb — then no counter can close it and a summons is coming. If a vehicle has been seized and is not being returned. If somebody was injured. If a licence has been taken under section 206(4) and the three months affects how you earn. Those are advocate matters and a drafted letter cannot reach them.

The directory costs nothing to use and nothing to send a request through. One request goes out to every advocate listed for that practice area at the same time, and the first one free picks it up. Your number stays hidden until somebody has accepted.

Court work is for your advocate, whose fee is engaged and paid by you directly; we do not quote, collect or share it.

What we charge

In-house drafting only, at one set of figures that does not change between jurisdictions.

An information request is five hundred rupees, and across this series it is the instrument that fits most often, because in about twenty of these places the question is simply whether a document exists. A reply in writing — for a figure above the Act’s ceiling, or a section that should never have been compounded — is two thousand five hundred. A plain application, for the ordinary asks, is one thousand five hundred. Searching the directory and sending a request through it cost nothing at all.

No part of an advocate’s fee comes to us.

And the one piece of advice that thirty-five jurisdictions have not shifted by a word: find the section, look up what the Act permits for it, pay what is owed, and file the receipt.

Questions people ask