Economic and industry statisticsEconomy and markets
Canadian new-home prices edge down 0.2% in September
The national index fell as Ontario and several other provinces declined, while British Columbia recorded a monthly increase.
- Event
- First published on EPC Feed
- Sources
- 3 listed below
- Reading time
- 3 min
Statistics Canada’s New Housing Price Index (NHPI) fell 0.2 per cent nationally in September, the agency reported on October 22. The index measures changes in agreed selling prices for new residential homes, comparing properties with the same detailed specifications across periods. It is not a measure of resale prices, home affordability, construction costs or the number of homes sold. The September result is not seasonally adjusted and is not subject to revision, according to the release.
Provincial results varied. Ontario declined 0.5 per cent month over month, while Saskatchewan fell 0.7 per cent, Newfoundland and Labrador declined 1.2 per cent, and New Brunswick was down 0.4 per cent. Alberta decreased 0.1 per cent, Nova Scotia 0.2 per cent, and British Columbia rose 0.5 per cent. Quebec, Manitoba and Prince Edward Island were unchanged. These movements show a national index masking different regional paths; the release does not attribute each province’s movement to a single factor.
The index focuses on prices agreed between a builder or contractor and buyer when a sales contract is signed. Prices are collected from builders and exclude value-added taxes such as GST and HST. The survey covers new single detached houses, semi-detached houses and townhomes. It provides results for Canada and the provinces, as well as 27 census metropolitan areas. Those design choices make the NHPI useful for tracking price movements in the new-home market while limiting what can be inferred: it does not cover every type of dwelling or every transaction in the housing market.
Statistics Canada’s technical guide explains that the series is designed to monitor changes in the price of newly built homes by holding detailed specifications constant between consecutive periods. The agency uses the index as one input in broader economic statistics, including parts of the Consumer Price Index and deflation of the national housing stock in the System of Macroeconomic Accounts. The NHPI is therefore both a housing-market indicator and a statistical tool for measuring real economic activity. It should not be confused with a builder’s cost index, which would track the cost of materials, labour or other inputs rather than the negotiated selling price.
The release states that coverage was expanded beginning with the August 2025 reference period as Statistics Canada incorporated additional data sources, and that this is an ongoing exercise. This provides context for interpreting the series over time: the index’s scope has evolved, even though the agency says the published index is not subject to revision. The technical documentation is important to users comparing trends across regions or relating new-home prices to other housing indicators.
The September figures arrived alongside other housing and construction data published during October, but they describe a different variable. Investment in building construction and building permits track spending or planned work; the NHPI tracks transaction prices for specified new residential structures. A fall in the index does not by itself mean fewer homes were built, nor does it prove that builders reduced costs. Similarly, provincial price changes do not measure the total price level or the total cost of purchasing a home in each province.
Statistics Canada scheduled its next NHPI release, for October 2026, for November 18. Until then, the September publication provides the latest official national and provincial comparison in this series. The national movement was modestly negative, but the underlying geography was mixed, with declines in several provinces and an increase in British Columbia. No single national result captures that full variation.
Sources
Spotted an error or have a primary source we should read? Send a correction.