The Surrey land share — the suburb conjecture, measured wrong

2026-07-19 · wave 28 · ESTIMATE lane, deepening mode · eighth city measured; closes the "City of Vancouver ≠ Vancouver CMA" caveat from wave 9 — in the opposite direction from the one we guessed.

What was measured

Surrey — BC's second-largest city (~600k people) and Metro Vancouver's biggest suburb — publishes its full property list as open data with BC Assessment's gross land and improvement values for every parcel (2026 year, 190,646 parcels). Aggregates via the city's ArcGIS statistics API, retrieved 2026-07-19:

Segment Parcels Land value Improvements Land share
All parcels, city-wide 190,646 $225.2B $61.9B 78.4%
Standard lots 103,557 $175.5B $46.2B 79.2%
Standard strata (condos) 84,722 $47.7B $15.4B 75.6%
Bareland strata 2,338 $2.0B $0.4B 84.3%
Largest single-family zone (R3) 59,508 $77.6B $15.6B 83.3%
One-acre residential (RA) 5,548 $17.2B $2.7B 86.6%

The correction (the fourth, and the most instructive)

Wave 9's caveat conjectured: "the CMA's suburbs (Surrey, Burnaby, Richmond) have lower land values and likely lower land shares." Half right, half wrong. Surrey's land values per parcel are indeed lower than Vancouver's — but its land share is higher: 78.4% city-wide against Vancouver's 75.4%, and the pattern holds segment by segment (standard lots 79.2% vs 74.8%; strata 75.6% vs 76.0%; single-family zones 83.3% vs 84.7%).

The conjecture assumed land share tracks land price. The data says it tracks supply constraint at the regional scale: Surrey sits inside the same mountain-ocean-ALR box as Vancouver, under the same regional scarcity, while its building stock is newer but cheaper per square foot of land. Scarcity is priced into every parcel of the region, not just the expensive core. The supply-constraint thesis (w21) doesn't just survive this measurement — it sharpens: the relevant unit is the metro land market, not the municipality.

What it changes

  1. The Vancouver incidence figures get more solid, not less. The −$49k/yr median-detached figure (NUMBERS #14) multiplies the CMA median price by a city-measured land share — the w9 caveat flagged that mix as the main vulnerability, guessing the suburbs would drag the share down. Measured: the biggest suburb's single-family-zone share (83.3%) sits within 1.4 points of Vancouver's (84.7%). The CMA-blended share is ~80–85%, and the headline stands, now on two measured municipalities covering the region's two largest cities.
  2. "Metro Vancouver is net-payer territory" now has a second data point — with a twist. Surrey's per-parcel land values are far below the city of Vancouver's ($1.69M average per standard lot vs $2.90M): a large share of Surrey's detached owners still sit near or below the $456k crossover, and its 84,722 condo units are squarely on the receiving side. High land share concentrates the schedule's weight on land-rich parcels without making a whole municipality a payer class.
  3. The BC–Quebec contrast is now four-vs-four. BC's four measurements (Vancouver 75%, Surrey 78%, Victoria 69% — plus w16's BC municipal cut) all sit far above Quebec's cluster (33–36%). Two provinces, two regimes, eight full rolls: the spectrum is a fact, and explaining it (real scarcity vs assessment convention — Q5) is the academic program's job.

Caveats

Sources

  1. City of Surrey Open Data, "Property Detail List" (ArcGIS FeatureServer, fields GROSS_LAND / GROSS_IMPROVEMENTS / GROSS_ASSESSMENT, LOT_TYPE, ZONE; current-year 2026 listing, item updated 2026-02-02), statistics queries retrieved 2026-07-19. https://services5.arcgis.com/YRpe0VKTJytZSSIB/arcgis/rest/services/property%20detail%20list/FeatureServer
  2. Vancouver comparison: analysis/vancouver-land-share.md (w9); Victoria: analysis/victoria-land-share.md (w19); Quebec cluster: analysis/quebec-rolls.md (w24/w26).
  3. Machine-readable rows: data/land-shares.csv.

Groundshare — a proposal in open development. Every number traces to a cited public source with its retrieval date; corrections are published, not erased. Rebuilt 2026-07-19 from the repo's research files.