Boundaries that don't match
Stand on a corner in New York City and you are simultaneously inside five non-aligned jurisdiction systems. The polygon fragmentation is one of the under-appreciated reasons NYC civic data is so hard to use, and so easy to misuse. To make the point: pick one block, and watch every system reach a different answer for it.
The exercise below uses 40 East 23rd St, Manhattan (Flatiron / Union Square area) as a fixed reference point — chosen because it sits at the intersection of multiple non-aligned systems. The marker doesn't move across beats; only the polygons under it change. Watch the highlighted shape on each beat: that's the feature this one address falls inside, in this one system.
Beat 1 · The system everyone knows
Boroughs are the system every New Yorker can name. Five of them, drawn in 1898 when the modern city was consolidated. Borough boundaries are old, political, and stable — they haven't meaningfully moved in over a century. The reference address sits in Manhattan, the borough that everyone gets right on the first try.
Everything that follows divides this same block into something else.
Beat 2 · Now zoom in
Neighborhood Tabulation Areas (NTAs) are NYC City Planning's 2020-revision boundaries — about 262 of them, drawn around Census-tract aggregations so that small-area population statistics actually publish. They are not the same as the neighborhoods residents name.
The reference address sits in Midtown South-Flatiron-Union Square. That hyphenated name is itself a hint: NTAs frequently merge two or three vernacular neighborhoods so the Census numbers don't get suppressed for small counts.
Beat 3 · Drawn in the 1970s
Community Boards are advisory bodies — there are 59 of them across the five boroughs, each composed of unpaid local volunteers appointed by the borough president. The polygons were drawn in the late 1970s under a Charter revision and have barely moved since. A CB hears complaints about traffic, sanitation, liquor licenses, and zoning variances; its formal power is limited but its political access is real.
The reference address sits in CD 105. The three-digit code starts with the borough (1 = Manhattan), so 105 reads as "Manhattan CB 5" — Flatiron, Union Square, the Madison Square area.
Beat 4 · Redrawn every ten years
New York City Council Districts are redrawn every ten years by the Districting Commission, tracking population shifts captured in each decennial Census. Fifty-one districts split the five boroughs into roughly equal-population units — which means their polygons drift with neighborhood-by-neighborhood growth, not with any of the older boundary systems.
The reference address sits in Council District 2. The Council Member for this district doesn't share borders with the Community Board chair, the State Assembly Member, or the police precinct's commanding officer — even though all four nominally cover "the same area."
Beat 5 · A 19th-century legacy
NYPD precinct boundaries date to the late 19th century, when they were drawn around the foot-patrol coverage radius of a beat cop. Seventy-eight of them survive, with periodic renumbering but rarely meaningful redraws — the geometry rewards continuity over coherence.
The reference address sits in the 13th Precinct (precinct 13). The precinct doesn't share a single edge with the Council District drawn around the same neighborhood. Two systems, both operational, both real. Neither one wins.
Borough · the reference address falls in Borough 1
One block. Five different answers.
The reference address is 40 East 23rd St, Manhattan (Flatiron / Union Square area). Here's what every system says it is.
- Borough
- Manhattan (code 1)
- NTA 2020
- Midtown South-Flatiron-Union Square (MN0501)
- Community District
- CD 105
- Council District
- 2
- Police Precinct
- 13
Why the fragmentation exists
Every jurisdictional boundary in NYC was drawn for a different reason at a different time, by a different body, in response to a different political pressure. Borough boundaries are old and political. NTAs are a 2020 Census-tract aggregation by City Planning. Community Boards are advisory and were drawn in the late 1970s. City Council Districts get redrawn every ten years per Charter. Police precincts are NYPD operational and were drawn for foot-patrol coverage in the 19th century. None of these processes were coordinated with any of the others. Each was internally rational. The collective result is a polygonal soup that no single agency owns.
Why it matters analytically
The fragmentation is what makes NYC civic data so hard to use. A 311 complaint includes the Community Board (CB) the complaint was filed in. A police incident includes the precinct. A school enrollment record includes the school district. A property tax record includes the Council District. Joining a 311 complaint to the Council District it falls into requires spatial intersection — taking the complaint's lat/long, finding the polygon that contains it, extracting the district number — because the dataset itself doesn't carry the join key.
Worse, the analytical errors that result from skipping the spatial join are subtle. If you aggregate 311 complaints by CB and report them as Council District statistics — using the one-CB-per-Council-District approximation that almost works — you produce numbers that look plausible but quietly miscount roughly 15-20% of records, because the boundaries don't actually align. The miscount can flip the rank order of districts in policy reports.
The civic-tech response
The most useful tool for working through the fragmentation is BetaNYC's Boundaries map, which visually overlaps all the major jurisdictional boundary systems on a single tiled basemap. You can click any address and see every administrative polygon it falls into, annotated with the relevant district numbers and the agency contacts. The tool exists because no city agency was willing to maintain the master cross-reference, and the civic-tech community filled the gap.
What the fragmentation actually means
NYC's polygonal soup is not a defect to be fixed. It is the visible consequence of a city governed by separate agencies that each evolved their own service-delivery geometry. To "fix" it would be to flatten 200 years of overlapping civic decision-making into a single grid, which would itself be politically catastrophic — your school district affects which school your kids can go to, your council district affects who represents you, and unifying the boundaries would require deciding whose existing boundary system wins. Easier to keep them all and build the cross-reference tools.
The deeper point is that the difficulty of doing analysis in NYC is partially a function of the city's governance complexity. The data is messier because the polity is messier. Open data made the complexity legible. Tools like Boundaries make it traversable. The polity itself remains complex by design.
Reference address memberships
40 East 23rd St, Manhattan (Flatiron / Union Square area)
coords: [-73.987, 40.74]
| System | Code | Label |
|---|---|---|
| Borough | 1 | Manhattan |
| NTA 2020 | MN0501 | Midtown South-Flatiron-Union Square |
| Community District | 105 | |
| Council District | 2 | |
| Police Precinct | 13 |