A study of the public record
How concentrated India's registered building actually is
There are hundreds of districts on the register. A handful of them hold most of it.
Every RERA filing names the district it sits in. Counted across all of them at once, the distribution is far more lopsided than a state-level view suggests: state totals hide the fact that a state's registrations are usually one or two districts deep.
This counts how much of the whole register sits in its largest districts, and lists them.
District is used rather than city deliberately. Several registers file a taluka or a ward in the city field — Maharashtra's records name Haveli and Mulshi where a reader expects Pune — so district is the only unit that means the same thing in every register.
Each bar: records, against the largest shown. Districts are scoped to their state, so two states' identically-named districts are never merged.
The 25 largest districts, and how much of the register they account for
| District | State | Records | Share | Running share |
|---|---|---|---|---|
| 1. Pune | Maharashtra | 12,831 | 10.3% | 10.3% |
| 2. Thane | Maharashtra | 6,906 | 5.5% | 15.8% |
| 3. Mumbai Suburban | Maharashtra | 6,271 | 5.0% | 20.8% |
| 4. Ahmedabad | Gujarat | 5,469 | 4.4% | 25.2% |
| 5. Raigarh | Maharashtra | 5,361 | 4.3% | 29.5% |
| 6. Bengaluru Urban | Karnataka | 4,545 | 3.6% | 33.2% |
| 7. Nashik | Maharashtra | 3,825 | 3.1% | 36.2% |
| 8. Ranga Reddy | Telangana | 3,540 | 2.8% | 39.1% |
| 9. Vadodara | Gujarat | 2,924 | 2.3% | 41.4% |
| 10. Palghar | Maharashtra | 2,684 | 2.2% | 43.6% |
| 11. Surat | Gujarat | 2,631 | 2.1% | 45.7% |
| 12. Medchal-Malkajgiri | Telangana | 2,540 | 2.0% | 47.7% |
| 13. Nagpur | Maharashtra | 2,494 | 2.0% | 49.7% |
| 14. Visakhapatnam | Andhra Pradesh | 2,044 | 1.6% | 51.3% |
| 15. Rajkot | Gujarat | 1,990 | 1.6% | 52.9% |
| 16. Kolkata | West Bengal | 1,986 | 1.6% | 54.5% |
| 17. Jaipur | Rajasthan | 1,853 | 1.5% | 56.0% |
| 18. Indore | Madhya Pradesh | 1,767 | 1.4% | 57.4% |
| 19. Gandhinagar | Gujarat | 1,659 | 1.3% | 58.7% |
| 20. Mumbai City | Maharashtra | 1,291 | 1.0% | 59.8% |
| 21. Sangareddy | Telangana | 1,261 | 1.0% | 60.8% |
| 22. Patna | Bihar | 1,250 | 1.0% | 61.8% |
| 23. Aurangabad | Maharashtra | 1,216 | 1.0% | 62.8% |
| 24. Guntur | Andhra Pradesh | 1,087 | 0.9% | 63.6% |
| 25. Krishna | Andhra Pradesh | 1,045 | 0.8% | 64.5% |
Shares are of records that name a district. The running share is cumulative down the table, largest first.
How this was counted
- Each record's filed district is read, normalised for case, and counted within its state. A district name is scoped to its state, so identically-named districts in two states are never merged.
- The cumulative column is the running share of all records carrying a district, largest first.
- Records with no filed district are excluded from the shares and counted separately.
What it does not show
- Coverage differs by register: a district that looks small may be a register Horizon holds only part of. Shares are of the corpus, not of every project registered in India.
- A small minority of filings carry the promoter's own district rather than the project's. In Karnataka, for example, around 4% of records name a district outside the state entirely, and their filed coordinates place the project inside Karnataka — so the district field there is an office address, not a site. Those records are counted where they were filed, which is the only thing the register actually says.
- District boundaries have been redrawn in several states during the period this data covers, and filings use whichever name was current. Two names can therefore describe overlapping ground.
- This counts registrations, not homes, floor area or money. A district of small plotted layouts can out-count a district of large towers.
- Concentration is a description of where filings are, not a judgement about any district, and not a signal about prices or demand in it.
Computed from every project record Horizon holds, sourced from the state RERA registers and recomputed daily. Horizon publishes the record; it does not rank, score or advise.
Cite this study
Free to quote, chart or republish with a link back to this page. No permission needed — the figures are counts of a public register. The numbers are recomputed daily, so a citation is worth dating.
43.6% of the register sits in ten districts, according to Horizon's count of India's state RERA registers (https://gethorizon.in/research/where-it-concentrates).
Horizon. “How concentrated India's registered building actually is.” Computed from India's state RERA registers. Retrieved —. https://gethorizon.in/research/where-it-concentrates