Building Construction Materials Price Indexes and Construction Price Indexes (Statistics Canada)

Building Construction Materials Price Indexes and Construction Price Indexes is Statistics Canada's construction-cost family: the official measurement of what it costs to build in Canada, tracked quarter by quarter since January 1981. The active flagship is table 18-10-0289-01, Building construction price indexes, by type of building and division - about 8,750 series totalling 310,204 datapoints, every value expressed on a 2023=100 base.

The cube is dimensioned three ways. Geography runs to 25 members: a fifteen-census-metropolitan-area composite at the top, then provinces and individual CMAs such as Toronto, Montréal, Vancouver, Calgary and Halifax. Type of building offers 16 members - residential buildings split into high-rise apartments, low-rise apartments, single-detached houses and townhouses; non-residential buildings covering commercial, industrial and institutional structures. Division contributes 24 members following UniFormat-style trade divisions: general requirements, demolition, concrete, masonry, structural steel framing and onward.

Around the flagship sits the rest of the family: annual relative-importance tables by type of building and division, and by type of building and census metropolitan area; an infrastructure construction price index; construction union wage rates and their index running monthly through June 2026; and investment-in-building-construction value tables linked from the same subject hub. Where it sits among its peers is mapped on our broadline retail data hub.

One row per reference period per geography per building type per division. Straight from the collection:

REF_DATE : 1981-01   GEO: Fifteen CMA composite
           Type of building: Non-residential buildings [622]
           Division: Division composite      UOM: Index, 2023=100
           VALUE: 24.6

REF_DATE : 1981-01   GEO: Fifteen CMA composite
           Type of building: Commercial buildings [62212]
           Division: Division composite      UOM: Index, 2023=100
           VALUE: 25.4

REF_DATE : 2026-04   GEO: Fifteen CMA composite
           Type of building: Residential buildings [621]
           Division: General requirements    UOM: Index, 2023=100
           VALUE: 110.3

The pair of 1981 rows is the pitch: non-residential construction at an index of 24.6 on today's base means the same bundle of building work costs more than four times what it did forty-five years ago, and the quarterly grid lets you watch that repricing arrive trade by trade rather than as one blended line. Request a sample and you get live rows in this shape for whichever geographies, building types and divisions you name.

Twelve verified fields form the dictionary, and they do two jobs at once. Four of them - REF_DATE, GEO, Type of building and Division - are the coordinates that locate every observation; because all four sit on each row, any cut of the panel is a filter rather than a restructure. The rest are bookkeeping that makes the panel dependable: UOM and UOM_ID pinning the index base, SCALAR_FACTOR stating the multiplier, VECTOR giving every series a permanent identifier, COORDINATE fixing its position inside the cube, VALUE carrying the reading itself, and STATUS/SYMBOL flags distinguishing preliminary from final figures and marking suppressions.

Companion-table measures beyond the flagship cube - relative importance, infrastructure indexes, union wage rates - fold under additional fields on request, pinned down against a delivered extract before anything depends on them.

Geography - Canada end to end. The fifteen-CMA composite gives one national-shape line, provinces give ten regional ones, and up to fifteen named census metropolitan areas give city-level readings including Toronto, Montréal, Vancouver, Calgary and Halifax. A national trend and a Toronto-specific high-rise cost curve come out of the same table.

Temporal - quarterly values from January 1981 through the current release; the 2026-04 reference quarter posted on 24 July 2026. Inactive historical vintages stretch further still, reaching back to 1956 for some utility and highway indexes, while the union wage-rate series run monthly through June 2026. Few economic series anywhere offer forty-five years of uninterrupted, consistently defined construction pricing.

Granularity - roughly 8,750 series formed as geography x building type x trade division, so the finest cut is something like high-rise apartment construction in Vancouver, concrete division. Annual relative-importance tables sit above that at division and CMA level, and monthly wage-rate indexes beside it. There is no project-level or firm-level detail anywhere in it - this is market-price data, not cost-estimate data.

API, files, or your warehouse. Daily, weekly, or hourly.

Your cadence is your call even though the underlying releases move once a quarter - load the 1981-onward history once as a backfill, then keep each new vintage rotating into place on whatever schedule your models expect. Samples ship in exactly the twelve-field shape shown above, cut to the geographies, building types and divisions you care about.

• Construction cost escalation and contract indexing - anchor escalator clauses and change-order formulas to a specific geography-building-type-division cell, so both sides of a contract resolve disputes against the same official series rather than a hand-collected number.
• Procurement and bid forecasting - read concrete, masonry and structural steel framing movements ahead of tender windows, using province and CMA cells to keep a national bid model honest about regional divergence.
• Investor and quant work - treat construction input inflation as an input-cost signal for building-products equities, homebuilder margins and REIT expense models, with forty-five years of history supporting real cycle analysis; see investors quants use cases.
• Real estate development feasibility - separate high-rise from low-rise from single-detached trajectories so pro formas reflect the building type actually being underwritten, not an all-construction average.
• Economic and policy analysis - cite the official measure of construction inflation behind deflators, capital-cost allowances and infrastructure program evaluation instead of reconciling unofficial estimates.

Data scientists and quant researchers (relevance 3 of 3) get a clean, fully keyed panel - four coordinates plus a permanent VECTOR identifier per series - that joins into cost models without fuzzy label matching; see data scientists use cases. Competitive intelligence and product teams (3 of 3) at building-materials suppliers and distributors get the official price environment their customers budget against, division by division. Investors and analysts (3 of 3) get input-cost context for construction-exposed portfolios across multiple full cycles. Market researchers and consultants (3 of 3) get the citable benchmark their construction-market deliverables are judged against. Journalists and academics get a named national statistical office standing behind every number, which makes attribution simple.

Within construction-cost data, this family owns the official Canadian whole-building view and concedes on materials-only granularity. The AGC Construction Data Hub tracks roughly 100 US producer-price series for individual materials - cement, lumber, steel mill products, gypsum - updated monthly, but stops at the US border and prices inputs rather than installed building work. The American Cement Association Market Intelligence goes deep on one material with forecasts, but one material only. The APA Engineered Wood Association data does the same for structural panels and engineered lumber.

What none of them replicates: whole-building installed cost, priced by trade division, across fifteen metropolitan markets and forty-five years, issued by the country's national statistical office. The usual pattern is to stack them - this family sets the authoritative Canadian baseline, the US materials feeds cover commodity input moves south of the border. The full slate sits on our best broadline retail datasets ranking.