The register was built to make projects finish, and for years it had no honest answer to a different problem: the project that should not start. Land locked in sudden litigation, a JV that collapsed, economics that died between registration and groundbreaking. Before 2023, such registrations just rotted into the lapsed lists, collecting enforcement flags. Order 42/2023 built a door, with one strict condition on who may use it.
Key takeaways
- Order 42/2023 (10 February 2023) created the deregistration route for projects that cannot proceed: litigation, unviability, funding failure, family or partner disputes, planning changes.
- The crucial prerequisite: no allottees, no bookings, nobody stranded by the exit.
- Almost nobody uses it. 500 published Maharashtra registrations read De-Registered against 14,003 Lapsed, twenty-eight lapses for every lawful exit.
- Complaints remain maintainable even after deregistration; the exit ends the project, not the accountability.
- A clean early deregistration beats a rotting lapse on every record that matters: the public filing, lender files and the promoter's own register history.
Why the door exists
A registration is a public promise with machinery attached: filing calendars, account structures, enforcement sweeps. When a project genuinely dies, that machinery keeps running against a corpse: missed QPRs draw show-cause notices, the registration lapses onto public lists, and the promoter's whole portfolio carries the flags into every future buyer's check and lender review, the same filed record MahaRERA's grading matrix was designed to score before it was shelved. The regulator recognised the waste on both sides: Order 42/2023 lets a promoter surrender the registration of a project that will not proceed, on stated grounds, litigation, economic unviability, funding, disputes, planning changes, provided the exit strands nobody.
How many promoters actually use it
The door has been open since February 2023, and the register records how often it is chosen over the alternative. The answer is: rarely.
Source: ReraGenie analysis of 55,913 published MahaRERA projects, as updated on 21 September 2026, counting each registration's filed current status. Completed (21,673) and Active (19,650) are excluded here because they are not endings of this kind.
Twenty-eight registrations lapse for every one that exits lawfully. Some of that gap is structural, since a lapse can happen to a project that is merely late while deregistration is only for one that has stopped, and a registration that lapsed in 2019 predates the door entirely. But the door has now been open for more than three years, and 500 uses against 14,003 lapses is not a rounding difference. Most promoters whose project died still let the registration rot rather than close it.
Where they do use it, the pattern is a young failure rather than an old one: 115 of the 500 were registered in 2023, more than any other year, and 14 in 2024. Geographically it tracks registration volume, Pune 148, Thane 64, Nagpur 62, Raigad 55, so nothing about the exit looks regional.
The zero-allottee condition, and why it is the whole design
The prerequisite that the project have no allottees or bookings is not a technicality; it is the entire moral architecture. Deregistration is an exit for empty projects, not a fire escape from sold ones: the moment buyer money entered, the promises attached, and the routes out run through refunds, consent and adjudication, never through surrendering the registration around the buyers. And the order's second guardrail seals the design: complaints remain maintainable even after deregistration, so whatever happened before the exit stays actionable after it.
That condition mostly holds where the filings can show it. Only 45 of the 500 deregistered projects file the building-level unit table at all, which is what a project that never started would look like, and 42 of those 45 record zero units sold. Three record a nonzero sold count, which the filings alone cannot explain: it may be a settled booking unwound before the exit, or a stale table nobody updated. Treat those three as a question rather than a finding, and treat the 42 as the shape of a proper deregistration: a registration that existed and a project that never reached a buyer.
An analogy: a company may wind up voluntarily only after settling creditors; it cannot dissolve its way out of debts. Deregistration is voluntary winding-up for projects, with allottees as the creditors whose absence, or full settlement, is the price of the door.
Running the exit properly
- 1
Decide early, on evidence
The moment the project's impossibility is established, litigation admitted, JV dissolved, feasibility dead, the calendar starts: every quarter of drift adds missed filings the exit cannot erase.
- 2
Confirm the zero-allottee state
No bookings, EOIs unwound with money returned, the paper trail complete. Any collection history needs resolving first, documented.
- 3
Apply on the stated grounds
The order's recognised categories, evidenced honestly: the application becomes part of the public record and reads accordingly.
- 4
Keep filing until the order issues
The registration lives until deregistration is granted; a compliance gap during the application undermines the story the application tells.
- 5
Close the loops after
Bank accounts wound down per the directions, land records cleared, and the exit noted accurately in future registration disclosures, where it will be read.
Source: MahaRERA Order 42/2023 and deregistration guidelines
The record arithmetic
Promoters resist deregistration as an admission, and the register prices the alternative worse. A project left to lapse accumulates: missed-filing flags, enforcement sweep appearances, a lapsed-list entry with the promoter's name, all of it read in every buyer's check and every counterparty file a landowner or lender ever builds. A clean deregistration is one line: registered, could not proceed for stated reasons, exited with nobody harmed. Iqbal's inherited portfolio (illustrative, as ever) carried one of each: the deregistered 2019 parcel draws a single question in lender meetings, answered in a sentence; the lapsed 2018 registration, dead the same year for similar reasons but never exited, still costs him a paragraph of explanation and a basis point or two of doubt, seven years on.
For landowners and JV partners, the deregistration record is a diligence lens in both directions: a counterparty with a clean exit has demonstrated they end things properly, and a counterparty whose failures all rotted into lapses has demonstrated the opposite. The lists tell both stories by name.
Reading any promoter's exit record, free
None of this needs a subscription to check. Every promoter on the register has a page listing their registrations with each one's current status, so a counterparty's ratio of finished to lapsed to deregistered is a minute's reading: start from the developer directory and open the entity your agreement will name, or from the district pages if you are working a market rather than a name. What the register will not tell you is WHY any single registration ended the way it did, which is exactly why the promoter who exited deliberately should expect to be asked, and should have the answer ready.
The exit record assembled the way a lender reads it, lapses and deregistrations across the promoter's registrations and around the project, with the filing record behind them, is part of what ReraGenie's Rs 2,999 project analysis puts in one document: your own record as the market sees it, or a counterparty's before their history becomes your disclosure.
The one-line summary
Order 42/2023 built the honest exit: empty projects only, stated grounds, accountability surviving the door. Twenty-eight registrations still lapse for every one that uses it, so using it early when a project dies is both the right thing and, on a register where the alternative is permanent, the cheap one.
Methodology and sources
- The deregistration route, its grounds, the zero-allottee prerequisite and the survival of complaints: MahaRERA Order 42/2023, dated 10 February 2023, and the deregistration guidelines issued under it.
- Status counts, registration years, districts and sold-unit figures: ReraGenie analysis of 55,913 published MahaRERA projects, as updated on 21 September 2026, reading each registration's own filed current status and, where filed, its building-level unit table.
- The 42-of-45 figure counts only deregistered projects that file that table; the other 455 file none, so the register cannot confirm their allottee position either way and this article does not claim it does.
- The register publishes a status and never a reason. Nothing here infers why a particular registration lapsed rather than exited, and a lapse can reflect a project that is merely overdue as easily as one that stopped.
Evaluating a micro-market or a land parcel?
The ReraGenie project analysis reads the filings around your parcel: supply, absorption and promoter records. Rs 2,999 per project; the area market report is a flat Rs 2,999 per pincode.
See the project analysis