SparkyData

NYC New Homes per Resident Compared with London, Paris, Toronto and Tokyo, 1990–2026: Permits, Starts, Completions and Net Additions, Each City Measured Its Own Way

Aggregate statistics from the article “On each city's own count of new homes, New York trailed London, the Paris region, Toronto and Tokyo per resident over the last decade”

No synthetic records. Published or computed tabulations containing no person-level records.

Records
Aggregate statistics
Place
New York City
Version
1.0,
Licence
CC BY 4.0; code MIT
Built for
On each city's own count of new homes, New York trailed London, the Paris region, Toronto and Tokyo per resident over the last decade (version 1.0)

From 2016 to 2025 New York permitted 2.8 homes a year per 1,000 residents, the lowest of five big cities on each one's own measure; Tokyo started 9.7.

What it shows

The article's own data file, aggregate statistics cut to what it analyses, and the script that computes every figure in it. Computed from: Census Bureau Building Permits Survey county files 1990-2025 and January-September 2025 and 2026; NYC Department of City Planning Housing Database (25Q4); MLIT Building Starts Statistics by prefecture, annual 1951-2025 and monthly to September 2026, and the Tokyo Metropolitan Government's starts by area 2015-2025; MHCLG Live Tables 118 (2000-01 to 2024-25) and 217 (1990 Q1 to 2026 Q2); SDES departmental estimates of dwellings authorised and started, January 2002 to September 2026 (vintage of October 2026), and its release of 28 October 2026; CMHC starts and completions via Statistics Canada Table 34-10-0154, 1972 to September 2026, and CMHC's City of Toronto starts 1998-2025; the City of Yes final plan; population estimates from the Census Bureau, the Tokyo Metropolitan Government, ONS, INSEE and Statistics Canada.

Files

Each file's columns are under Columns below and in data_dictionary.csv.

FileRowsSizeWhat it holds
data.csv 380 12 KB The data
analysis.py 24 KB The analysis script

Columns

data.csv: The data (4 columns)

The aggregate statistics the article's figures are computed from.

measure, year, homes, per_1000. The article and its script define each one.

Licence and credit

Data CC BY 4.0; the script MIT. Records we counted from government files stay under their publishers' terms. Public-use microdata are disclosure-protected records derived from survey responses, published by the issuing agency so that no respondent can be identified; synthetic records are generated by SparkyData and correspond to no real person, household or address. Neither may be used to identify anyone.

Credit it as:

SparkyData (2026). NYC New Homes per Resident Compared with London, Paris, Toronto and Tokyo, 1990–2026: Permits, Starts, Completions and Net Additions, Each City Measured Its Own Way (version 1.0). Data CC BY 4.0.
https://sparkydata.dev/data/nyc-new-homes-per-resident-five-big-cities
Sources (8)

The article gives the charts, the uncertainty of each comparison and its own changelog. Corrections and questions: sparkytech.dev@gmail.com, or see corrections.