Housing Statistics in Canada
A toolkit for understanding housing supply

By Florian Mayneris and Radu Andrei Pârvulescu

Release date: October 25, 2023

Skip to text

Text begins

Overview

This article draws together a variety of publicly available data sources into a toolkit of indicators that can be used by researchers, practitioners and the public to describe housing dynamics from a supply-side lens. The article then leverages this toolbox to illustrate trends in housing prices, supply and key determinants of residential construction over the past 12 years in selected census metropolitan areas (Halifax, Montréal, Ottawa–Gatineau, Toronto, Edmonton and Vancouver). The purpose of this article is both to highlight the rich variety of public data sources available for supply-side analysis and to mark out areas where data gaps exist.

Highlights

  • New data on vacant land show that the census metropolitan areas of Ottawa–Gatineau (Ontario part), Halifax and Toronto have more residentially zoned vacant land than Edmonton or Vancouver. In Toronto and Vancouver, almost all vacant land is located outside the CMA core.
  • Building construction costs have increased in major Canadian cities, with a clear acceleration since the COVID-19 pandemic.
  • Average wages in the construction industry have increased less than in the rest of the economy, except in Nova Scotia.
  • The job vacancy rate for the building trades has rapidly increased in all provinces since the COVID-19 pandemic.
  • Rich data are available on key components of housing supply—from the existence of residential vacant land to the availability of workers in the construction sector and the costs of building materials.
  • Data gaps exist with regard to land assembly, zoning and permitting, land banking, development costs, and inspection and certification.

Introduction

Housing prices have increased in several Canadian cities, especially during the COVID-19 pandemic. Since the second quarter of 2022, housing unaffordability in Canada has reached levels not seen since the early 1990s (Bank of Canada, 2022). While 2023 has seen house prices decrease in some areas, housing affordability remains a major concern to both citizens and the various levels of government.

Housing prices and construction costs are jointly determined by the demand for and supply of housing, be it for use as lodging or as an investment vehicle. In the recent context of deteriorating housing affordability in Canada, insufficient housing supply has been proposed as one of the possible drivers of rising housing prices, sparking renewed interest in the determinants of housing supply (see, for example, CMHC, 2022). Academic research further highlights the “combination of strong and growing demand for housing in desirable areas in conjunction with tight long-term supply constraints” (Hilber and Schöni, 2022).

This article begins with the context of recent trends in housing prices, and then describes housing supply and several factors important for residential construction, for selected Canadian cities. One important takeaway is that conditions vary widely across the country, with few truly national trends. The article draws together a variety of publicly available data sources into a toolkit of indicators that can be used by researchers, practitioners and the general public to describe housing dynamics from a supply-side lens. The present article also points at data gaps that prevent drawing a fuller picture of housing supply. It is hoped that this toolkit will facilitate analysis of the numerous factors affecting housing supply in various Canadian cities and will contribute to identifying avenues for expanding the range of indicators available to stakeholders, citizens and researchers.

All indicators are presented on a yearly basis from 2010 onwards for six census metropolitan areas (CMAs): Halifax, Montréal, Ottawa–Gatineau, Toronto, Edmonton and Vancouver. Together, these CMAs contain almost half (44.6%) of the Canadian population, according to the 2021 Census of Population.Note That said, the frequency and geographic coverage of these indicators vary significantly; see Appendix A for details and links.Note

1 Housing price context

House prices are central to housing affordability, since the cost of acquiring a home (through a purchase or renting) is typically the largest component of shelter costs, which include elements—such as utilities and interest payments on a mortgage—that are not considered here. Moreover, the selling price of a house is important for property developers and builders, who typically enter a housing market only when the price is high enough to accommodate their desired profit margin. It is therefore important to first survey the evolution of several housing price indicators, as this will give context in which other measures—such as the volume of new dwelling units or changes in the price of construction materials—may be interpreted.

Four publicly available indicators capture the evolution of housing prices in different segments of the housing market. The Teranet–National Bank House Price Index (TN-HPI) uses information from land registries and a repeat sales methodology to capture price changes in the existing stock of properties. The New Housing Price Index (NHPI), provided by Statistics Canada (StatCan), captures selling prices for new residential houses via a questionnaire sent out to a sample of residential builders.Note The New Condominium Apartment Price Index (NCAPI), also produced by Statistics Canada, complements the NHPI with information on developers’ selling prices of units in new condominium apartment buildings. Finally, price changes in the rental market can be tracked via the Rental Market Survey (RMS) from the Canada Mortgage and Housing Corporation (CMHC).Note Note

Chart 1 below shows that the yearly evolution of these four indicators of housing prices—the TN-HPI (existing houses), the NHPI (new houses), Note the NCAPI (new condo apartments)Note and the RMS (rents)—are highly correlated, with pairwise correlations above 70%. Chart 1 confirms that these four indexes paint a consistent picture. There is an overall trend of price increases over the past 12 years in all six CMAs considered here, though both the magnitude and the timing of this evolution varied across cities.

Prices strongly trended upwards from 2010 to 2022 in Toronto (+184.8% for the TN-HPI, +50.9% for the NHPI, +39.1% for the NCAPI and +58.4% for the RMS) and Vancouver (+129.9% for the TN-HPI, +33.1% for the NHPI, +21.8% for the NCAPI and +67.1% for the RMS), compared with other CMAs.

Resale house prices (TN-HPI) increased to a lesser extent in Halifax, Montréal and Ottawa–Gatineau. In these three cities, the rise in price indicators was concentrated from 2018 to 2022.

Finally, compared with the other five CMAs, prices were more stable in Edmonton (+15.5% for the TN-HPI, +11.9% for the NHPI, -12.3% for the NCAPI and +27.7% for the RMS).

Regarding rents, slower price growth partly reflects the fact that many jurisdictions have rules and regulations governing the amount of rent increases that landlords can impose on tenants. It is also worth noting that the price of existing homes (the TN-HPI) increased faster than the other three indexes, except in Edmonton. This does not imply that existing homes are generally more expensive than new builds, but rather that prices for existing stock have appreciated quicker. This may occur when rapidly increasing housing demand outpaces developers’ ability to build and sell new houses.

Chart 1

Data table for Chart 1 
Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1. The information is grouped by CMA (appearing as row headers), NCAPI, NHPI, RMS and TN-HPI, calculated using index units of measure (appearing as column headers).
CMA NCAPI NHPI RMS TN-HPI
index
Edmonton
2010 Note ..: not available for a specific reference period 100 100 100
2011 Note ..: not available for a specific reference period 100.9 101.7 98.4
2012 Note ..: not available for a specific reference period 101.9 105.3 100.2
2013 Note ..: not available for a specific reference period 102.3 112.2 103.5
2014 Note ..: not available for a specific reference period 102.4 120.6 108.5
2015 Note ..: not available for a specific reference period 102.8 123.6 110.6
2016 Note ..: not available for a specific reference period 102.6 120.8 109.6
2017 100 102.3 119.2 109.5
2018 97.2 101.9 122.2 109
2019 92.9 100.6 123.3 107
2020 88.3 99.9 124.5 105.7
2021 87 104.5 124.6 110.7
2022 87.7 111.9 127.7 115.5
Halifax
2010 Note ..: not available for a specific reference period 100 100 100
2011 Note ..: not available for a specific reference period 101.7 103.8 103.1
2012 Note ..: not available for a specific reference period 103.9 107 109.6
2013 Note ..: not available for a specific reference period 106.6 109.5 110.6
2014 Note ..: not available for a specific reference period 106.9 112.8 108.9
2015 Note ..: not available for a specific reference period 107.7 117.6 109.6
2016 Note ..: not available for a specific reference period 108.1 119.3 108.9
2017 100 109.5 124.5 111.5
2018 104.4 110.7 129.7 115.2
2019 107.7 111.5 134.9 120.8
2020 111.5 115.6 140.7 135.1
2021 122.8 126.6 149.7 174.6
2022 119.9 131.6 162.6 212
Montréal
2010 Note ..: not available for a specific reference period 100 100 100
2011 Note ..: not available for a specific reference period 102.9 102.6 105.8
2012 Note ..: not available for a specific reference period 104.3 101.4 110.3
2013 Note ..: not available for a specific reference period 105.3 104.1 111.6
2014 Note ..: not available for a specific reference period 105.7 105.4 114.2
2015 Note ..: not available for a specific reference period 106.1 108.4 115.9
2016 Note ..: not available for a specific reference period 107.2 112.8 117.6
2017 100 108.4 111.6 122.1
2018 102.9 111.2 115.4 127.8
2019 106.1 115.5 122.1 135.4
2020 113.3 124.6 128.8 150.5
2021 121.4 145.8 133 177.1
2022 119.6 161.8 145.8 198.5
Ottawa-Gatineau
2010 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 100 100
2011 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 103.2 105
2012 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 104.8 108
2013 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 106.5 109.4
2014 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 107.2 110.1
2015 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 110.2 109.7
2016 Note ..: not available for a specific reference period Note ..: not available for a specific reference period 112.1 111.8
2017 100 100 114.7 115.7
2018 101.8 104.2 120.2 121.4
2019 117.4 109.1 131.4 129.4
2020 134.6 119.8 138.9 148.6
2021 138.5 142.4 145.6 180.7
2022 138.6 156.1 158.3 198.1
Toronto
2010 Note ..: not available for a specific reference period 100 100 100
2011 Note ..: not available for a specific reference period 104.7 102.2 106.9
2012 Note ..: not available for a specific reference period 110 105.3 115.9
2013 Note ..: not available for a specific reference period 112.7 107.8 121
2014 Note ..: not available for a specific reference period 115.1 111.4 129.1
2015 Note ..: not available for a specific reference period 118.8 114.5 140.1
2016 Note ..: not available for a specific reference period 126.3 118.1 160.4
2017 100 135.5 124.9 192.8
2018 106.2 135.3 130.7 195
2019 111.2 134.1 139.2 202.2
2020 128.6 135.3 145.8 219.5
2021 133.1 144.4 149.6 253.5
2022 139.1 150.9 158.4 284.8
Vancouver
2010 Note ..: not available for a specific reference period 100 100 100
2011 Note ..: not available for a specific reference period 99.7 103.4 107.9
2012 Note ..: not available for a specific reference period 99.2 105.4 109.4
2013 Note ..: not available for a specific reference period 98.1 107.1 109.8
2014 Note ..: not available for a specific reference period 97 109.6 117.2
2015 Note ..: not available for a specific reference period 97.9 114.4 127.4
2016 Note ..: not available for a specific reference period 102.5 121.4 154.1
2017 100 108.8 129.6 171.8
2018 104 112.8 137.4 187.7
2019 106.4 111.2 145.8 179.4
2020 110.7 112.7 149.7 184.7
2021 112.7 125 152.2 209.7
2022 121.8 133.1 167.1 229.9

2 Data on housing supply and related factors

Housing construction costs and the volume of new residential builds are jointly determined by demand for and supply of housing, whether for use as lodging or as an investment vehicle. Housing demand is outside the scope of this article, which instead focuses on some of the determinants of the supply of new dwellings in selected CMAs. Before discussing trends, however, it is worth providing an overview of what can be measured with publicly available data and identifying where the data gaps are.

2.1 A wide range of publicly available data…

Determinants of housing supply

The development of new properties can be conceptualized as a sequence of different steps, each of which involves various factors that matter for the supply of new dwellings. Figure 1 provides a visual representation of these different steps and the way they will be discussed in the following section.

Figure 1

Description for Figure 1

The title of the figure is “Sequencing of housing development, with associated data sources”. This figure shows a four-step flow chart of the process of developing residential housing. Each step is associated with data sources that can be used to describe that portion of the process. Associated data sources come from Statistics Canada (StatCan) or from the Canada Housing and Mortgage Corporation (CMHC). Certain data sources are missing, representing data gaps to be filled by future research.

The steps in the figure are:

  1. Secure buildable land
  2. The associated data sources are CHSP and MLUR (from StatCan), and Land assembly and Zoning (Data gaps)

  3. Decide to develop land (with three sequential components):
    1. Developer requests permit
    2. Municipality issues permit
    3. Developer decided to build

    The associated data sources are MLUR and Building permits (from StatCan), Housing Starts (from CMHC), and Land banking and Development costs (Data gaps)

  4. Construct
  5. The associated data sources are BCPI, IPPI, Job vacancies, Wages, Apprenticeships, and Employment (from StatCan)

  6. Inspect and certify new builds
  7. The associated data sources are Housing completions (from CMHC) and Inspections and certifications (Data gaps)

The first step in providing new housing supply (as described in Box 1) is to secure buildable land, namely the series of activities from ensuring that a plot is available, residentially zoned, and conforming with density limits and parking requirements, among other regulations. For the first time, the Canadian Housing Statistics Program (CHSP) at Statistics Canada is releasing preliminary information on residential vacant land in terms of the land area (in acres) that these vacant lots cover. The CHSP remains the first and only national source of information on the quantity of land readily available for residential construction in Canadian cities. Moreover, in the spirit of the Wharton Residential Land Use Regulation Index for U.S. cities, comprehensive data on local housing regulations—for example, on zoning and density requirements—have recently been collected in a joint initiative between Statistics Canada and CMHC for the largest Canadian CMAs, namely the Municipal Land Use and Regulation (MLUR) survey. Like the Wharton index, these new data are based on surveys of local governments and focus on their perception of processes and outcomes. A first analysis of MLUR data has already been undertaken by CMHC and reveals a higher regulatory burden in Vancouver and Toronto, cities also marked by high housing demand and a greater incidence of high-rise developments.

The second box concerns the decision to develop the land, a joint process between a property developer and municipal authorities. The developer starts by requesting a building permit; the municipality reviews and approves, rejects or suggests modifications to the application; and then the developer decides whether it will delay or begin construction, or sell the still-vacant plot. Three available sources of data provide information on this step. Information on building and demolition permits issued at the CMA level is available, where building permits measure intended construction projects approved by local authorities—as opposed to housing starts and construction investment that measure actual construction activity.Note Housing starts are estimated via a CMHC survey that either asks builders about the state of construction projects or conducts on-site visits.Note The MLUR survey mentioned above has also collected data on the process required to obtain development approvals or building permits.

Box 3 deals with the construction process, with available indicators (all from Statistics Canada) particularly focused on construction costs. The Building Construction Price Index (BCPI), in its residential part, measures change over time in the prices that contractors charge to construct residential buildings. It aggregates several dimensions of construction costs, some of which can be tracked with publicly available data.Note Alternatively, for construction materials, the Industrial Product Price Index (IPPI) measures price changes for major commodities sold by manufacturers operating in Canada.Note While this is only an approximation of the price paid by the construction industry, the IPPI may nonetheless provide useful insights on the evolution of prices for specific building construction materials.Note Data on the labour market for the construction trades provide information on construction costs via the availability of workers and wages. Data are available on the number of both apprentices and workers in the building trades, as well as on the vacancy rate of construction-industry jobs and the average hourly wages offered in this sector.Note These indicators are available at the provincial level.Note

The fourth and final box is on the completion and on the inspection and certification (e.g., to fire code) of new buildings. In the same survey as the one providing housing starts, CMHC provides estimates of the number of completed residential construction units (housing completions).

These data sources form the housing data toolkit at the base of this article’s findings. Full descriptions of the data in terms of geographies available and time grain, as well as links, can be found in Appendix A.

2.2 …but also significant data gaps

Although the available data on housing prices and supply are rich, data gaps remain for several dimensions of the housing supply process identified in the literature. These gaps appear especially before and after the construction phase of the development process. At the beginning, there is a lack of data on when property developers assemble land, potentially apply for zoning, seek permits, begin or delay construction, or sell the plot. At the end of the process, data are missing on the inspection and certification process of newly built residential units. Regarding the securing of developable land, a first gap is land assembly. Developers often need to join adjacent parcels of land that are too small to host a construction project. Holdouts—i.e., property owners refusing to sell the parcels necessary for particular building projects (Brooks and Lutz, 2016)—may increase the cost of land, as well as the administrative costs of building projects. Little is known about land assembly and holdouts since tracking mergers and divisions of parcels requires tracking historical land titles over time.

The concrete application of municipal regulations is another area with insufficient data. Notwithstanding the recent data collection efforts undertaken by Statistics Canada and CMHC about perceptions of existing regulations, there remains a lack of information on how local development regulations are implemented—for instance, on the frequency with which the rezoning applications of developers are approved. There is also a relative lack of data on development costs, a common mechanism whereby municipalities charge developers for anticipated capital upgrades in the public infrastructure needed to support new housing, such as larger storm drains or expanded fire services.Note

Information about land banking is also sparse. Land banking may occur when developers have acquired land but do not use it, as they wait for market conditions that will make it more profitable either to start building or to resell the parcels (Murphy, 2018; Murray, 2020). No publicly available data currently allow the extent of this phenomenon in Canadian cities to be measured.

Finally, regarding building inspections and certifications, new or renovated buildings may be inspected several times during construction to ensure they meet building codes. Often, residential buildings must receive a permit of occupancy before they can be inhabited. There is little research on the inspection and certification process, e.g., the associated costs and timeframes, and the frequency of delays.

Data gaps at the beginning of the development process are especially salient because public policy often aims to incentivize developers early on to make construction more likely. Public authorities may, for example, offer concessions from zoning rules, or property swaps with public lands.Note Assessing the effectiveness of such policies therefore requires high-quality, longitudinal data on changes in land parcel boundaries, zoning (both as it appears in local regulations and as it is actually applied), permitting, or the use of developable land (in particular whether and how long it is kept empty).

3 Recent trends in housing supply and its determinants

3.1 Housing construction and investment

In terms of construction activity, the investment in building construction (IBC) indicator, produced by Statistics Canada, captures the value spent by households, enterprises and governments for the construction of buildings, excluding the value of land. This paper presents the residential part of the IBC indicator (IBC-R), which is based on building permits, starts and completions, and administrative data that adjust the base value of construction investment to account for industry profit and other costs not normally included in the value of a building permit. Therefore, although all three indicators measure different things (e.g., not all building permits end up in new constructions), IBC-R , building permits and housing starts are highly correlated, as illustrated in Chart 2.Note Hence they are here considered alongside each other.

It is more difficult to discern trends in housing construction and investment than in the housing prices surveyed earlier. Vancouver is the only CMA in which housing construction indexes exhibited a coherent upward trend over the past 12 years (+97.6% for the number of new dwellings in building permits, +70.3% for housing starts and +124.9% for the IBC-R ). In Halifax (+85.3% for the number of new dwellings in building permits, +38.4% for housing starts and +144.6% for the IBC-R ) and Ottawa–Gatineau (+45.8% for the number of new dwellings in building permits, +63.8% for housing starts and +140.9% for the IBC-R ), the supply of new dwellings has also increased, albeit largely from 2016 onwards. Toronto (+31.0% for the number of new dwellings in building permits, +54.1% for housing starts and +93.5% for the IBC-R ) and Edmonton (+49.6% for the number of new dwellings in building permits, +44.7% for housing starts and +25.6% for the IBC-R ) also experienced an overall increase in the number of new housing units, but this growth is less pronounced and more irregular over the period under consideration. Finally, housing construction does not exhibit clear trends in Montréal, where construction also increased less than in the other five CMAs, especially when measured in terms of the number of permitted dwellings (-17.5%) and housing starts (+9.9%).

Chart 2

Data table for Chart 2 
Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2. The information is grouped by CMA (appearing as row headers), IBC-R, Housing starts and Number of building permits (index), calculated using index units of measure (appearing as column headers).
CMA IBC-R Housing starts Number of building permits (index)
index
Edmonton
2010 100 100 Note ..: not available for a specific reference period
2011 95 92.6 100
2012 107.3 128.3 122.8
2013 121.3 146.5 136.5
2014 141 138.7 160.3
2015 145.3 170.8 164.4
2016 106.4 100.2 112
2017 91 113.6 117.6
2018 102.1 100.5 113.7
2019 87.3 106.7 101.2
2020 84.9 115.7 117.9
2021 110 125.5 130.5
2022 125.6 144.7 149.6
Halifax
2010 100 100 Note ..: not available for a specific reference period
2011 109.9 121.2 100
2012 121.5 113.8 98.3
2013 115.1 99.7 67.1
2014 102.9 72.4 70.8
2015 122.4 106.1 89.2
2016 115.3 94 78.2
2017 136.1 112.2 96.1
2018 146.8 117.8 100.4
2019 172.3 128.4 133.4
2020 170 134.7 124.9
2021 184.9 158.5 154.6
2022 244.6 138.4 185.3
Montréal
2010 100 100 Note ..: not available for a specific reference period
2011 103.1 102.9 100
2012 107.4 93.7 87.4
2013 99.5 70.7 74.6
2014 100.2 84.5 75.2
2015 101 84.9 70.7
2016 109 80.5 79.3
2017 125.2 112 104
2018 147.1 113.2 129.3
2019 155.4 113.6 107.3
2020 147.4 124.2 118.8
2021 193.5 147.7 119.1
2022 184.7 109.9 82.5
Ottawa-Gatineau
2010 100 100 Note ..: not available for a specific reference period
2011 105 89.7 100
2012 99.8 96.5 100.1
2013 107 91.8 77.5
2014 129.8 83.6 98.5
2015 126.4 70.5 67.3
2016 128.8 76.7 97.9
2017 146.1 102.1 97.6
2018 161.4 102.8 117.9
2019 177.2 121.6 144.8
2020 229.4 142 160.4
2021 232 146.7 155.2
2022 240.9 163.8 145.8
Toronto
2010 100 100 Note ..: not available for a specific reference period
2011 104.8 134.7 100
2012 113.7 163.5 116.2
2013 117.6 114.5 120.7
2014 123.8 97.9 104.7
2015 128 142.7 119.1
2016 143.7 132.5 114.6
2017 148.5 133.4 101.6
2018 141.3 140.6 125.7
2019 145.2 103.3 110.3
2020 150.8 131.3 145.4
2021 186 142.6 178.8
2022 193.5 154.1 131
Vancouver
2010 100 100 Note ..: not available for a specific reference period
2011 107.3 117.1 100
2012 111 124.8 107.3
2013 122.3 122.7 114.9
2014 127.9 126.1 108.3
2015 139.7 136.9 145.3
2016 163.3 183.2 130.5
2017 164 171.6 152.1
2018 181.2 153.9 158.9
2019 195.7 184.5 154.7
2020 178.4 146.7 133.4
2021 184 170.8 145.7
2022 224.9 170.3 197.6

3.2 Vacant land

Securing land is a crucial step in a housing development project. This is done either by demolishing existing structures to make way for new housing or by obtaining a plot of vacant land. The latter is typical for developers seeking to build subdivisions, and the use of vacant land for residential construction has been at the centre of debates on both housing affordability and urban expansion. It is therefore important to be able to measure how much vacant land is immediately available for construction.

Chart 3 shows the residential vacant land area (in acres) as a share of a CMA’s total land area for 2021. The Ottawa–Gatineau CMA (Ontario part) had the highest quantity of residential vacant land, both as a share of the CMA’s overall land area (18.0%), and in terms of total acreage (162,000 acres). Land availability may therefore be less of a constraint on the housing supply in this CMA than that in Halifax (11.1% of CMA land area, 151,000 acres), Toronto (9.0%, 131,000), Vancouver (4.5%, 32,000) and Edmonton (2.5%, 59,000). These percentages should be interpreted with care, however, since the total surface area of CMAs varies widely. The CMA of Edmonton, for example, has a land area of 9,439 km2, compared with 2,883 km2 for Vancouver. It is therefore not surprising that residential vacant land lots should account for less of the total in the former, despite there being more acres of residential vacant land in Edmonton than in Vancouver.Note

It is also possible to analyze the share of vacant land area in the core city of a CMA (such as the City of Toronto), Note as opposed to the rest of the CMA, which often has a suburban or rural character.Note Chart 4 reveals that a high percentage of vacant land area in both the Ottawa–Gatineau (Ontario part) (75.9%) and Edmonton (33.9%) CMAs was located within the limits of the core city in 2021, while the proportion in the cores was much more limited in the Toronto (2.1%) and Vancouver (0.3%) CMAs. The constraint of vacant land availability in the cores of Toronto and Vancouver is thus stronger than suggested by CMA-level statistics. These figures show that even if a CMA features large stocks of residentially zoned vacant land, these stocks may be almost entirely outside the CMA’s core city.Note This may have differing implications for housing supply, as it pertains both to the relative ease of developing previously built-over land (“greyfill”), as opposed to previously unbuilt parcels (“greenfill”), and to housing affordability.

Beyond the availability of vacant land, the ownership structure of those properties may also matter for housing supply. Additional information on the characteristics of vacant land is publicly available, such as whether vacant land is owned by a person or by a business or government. For example, a recent CHSP article reveals that in the Atlantic provinces, vacant land was more often owned by people who owned one or two pieces of vacant land in addition to their primary place of residence, compared with other provinces (Fontaine and Gordon, 2023).

Chart 3

Data table for Chart 3 
Data table for Chart 3
Table summary
This table displays the results of Data table for Chart 3. The information is grouped by CMA (appearing as row headers), Percentage of the total area covered by residential vacant land in 2021, calculated using percent units of measure (appearing as column headers).
CMA Percentage of the total area covered by residential vacant land in 2021
percent
Edmonton 2.5
Halifax 11.1
Ottawa-Gatineau 18.0
Toronto 9.0
Vancouver 4.5

Chart 4

Data table for Chart 4 
Data table for Chart 4
Table summary
This table displays the results of Data table for Chart 4. The information is grouped by CMA (appearing as row headers), Percentage of residential vacant land area situated in the core city in 2021, calculated using percent units of measure (appearing as column headers).
CMA Percentage of residential vacant land area situated in the core city in 2021
percent
Edmonton 33.9
Halifax Note ..: not available for a specific reference period
Ottawa-Gatineau 75.9
Toronto 2.1
Vancouver 0.3

3.3 Building construction costs

The evolution of building construction costs has been qualitatively homogeneous across cities in the past 12 years. Chart 5 show that overall residential construction costs, as measured by the BCPI, have increased in the six CMAs, accelerating notably from 2020. This partly reflects supply-chain bottlenecks caused by the COVID-19 pandemic (Meyer-Robinson, 2022), as well as tightening labour shortages in the construction sector (Morissette, 2022). While the BCPI’s evolution is qualitatively similar across the CMAs, the size of its growth varies, with Toronto (+74.0%), Ottawa–Gatineau (+69.4%) and Edmonton (+62.2%) standing out.

It is also notable that the price of key construction materials (as captured by the IPPI) has surged since the start of the pandemic, in particular that of lumber (+40.0% in 2020 and +61.7% in 2021), prepared asphalts (+7.9% in 2021 and +24.4% in 2022) and ready-mixed concrete (+11.5% in 2022).

Chart 5

Data table for Chart 5 
Data table for Chart 5
Table summary
This table displays the results of Data table for Chart 5. The information is grouped by Year (appearing as row headers), Edmonton , Halifax, Montréal, Ottawa-Gatineau, Toronto and Vancouver, calculated using index (2017 = 100) units of measure (appearing as column headers).
Year Edmonton Halifax Montréal Ottawa-Gatineau Toronto Vancouver
index (2017 = 100)
2017 100.0 100.0 100.0 100.0 100.0 100.0
2018 104.5 105.2 103.2 106.6 107.8 107.7
2019 105.9 108.0 107.4 112.0 109.9 112.8
2020 110.2 112.2 111.9 118.5 114.1 116.1
2021 133.7 127.0 127.3 147.0 139.0 127.2
2022 162.2 143.9 146.8 169.4 174.0 144.9

3.4 Labour market determinants of housing supply

Housing supply is partly influenced by the labour market in the construction industry. Of particular importance are the availability of labour (both current workers and apprentices in the construction trades) and the level of wages in the construction sector.

Regarding the number of workers, Chart 6 shows that from 2010 to 2022, the growth rate of employment in construction was highest in Ontario (+34.3%) and Quebec (+26.7%). In terms of training, the number of apprentices remained largely stable (in British Columbia and Ontario) or decreased (-35.1% in Alberta) from 2010 to 2021, except in Nova Scotia and Quebec (where it increased by 31.9% and 31.2%, respectively).

Chart 6

Data table for Chart 6 
Data table for Chart 6
Table summary
This table displays the results of Data table for Chart 6. The information is grouped by Province (appearing as row headers), Employment and Number of apprentices, calculated using index units of measure (appearing as column headers).
Province Employment Number of apprentices
index
Alberta
2010 100.0 100.0
2011 110.3 94.2
2012 117.1 94.3
2013 116.0 101.2
2014 124.0 110.5
2015 126.0 108.9
2016 120.0 99.0
2017 115.4 86.2
2018 116.8 77.4
2019 116.2 70.4
2020 108.7 61.0
2021 113.0 64.9
2022 118.3 Note ..: not available for a specific reference period
British Columbia
2010 100.0 100.0
2011 102.6 95.7
2012 102.7 95.7
2013 107.7 97.9
2014 105.9 101.9
2015 107.3 103.2
2016 111.4 102.6
2017 121.5 102.2
2018 124.9 105.5
2019 126.7 106.9
2020 115.2 103.1
2021 115.0 104.7
2022 119.6 Note ..: not available for a specific reference period
Nova Scotia
2010 100.0 100.0
2011 101.2 107.3
2012 95.3 114.7
2013 102.6 118.5
2014 102.3 118.4
2015 102.9 115.8
2016 97.4 120.0
2017 93.0 125.0
2018 90.9 122.0
2019 98.8 127.3
2020 96.2 128.7
2021 105.0 131.9
2022 124.0 Note ..: not available for a specific reference period
Ontario
2010 100.0 100.0
2011 102.4 97.9
2012 106.0 103.3
2013 107.4 112.1
2014 107.0 97.2
2015 108.9 102.2
2016 113.3 90.1
2017 116.9 89.7
2018 120.6 84.6
2019 124.3 88.7
2020 121.3 85.4
2021 123.0 94.6
2022 134.3 Note ..: not available for a specific reference period
Quebec
2010 100.0 100.0
2011 108.7 102.9
2012 107.2 105.0
2013 115.9 109.5
2014 108.7 102.3
2015 101.7 101.3
2016 102.2 101.4
2017 108.4 102.9
2018 106.7 108.7
2019 113.4 117.0
2020 113.0 120.4
2021 122.0 131.2
2022 126.7 Note ..: not available for a specific reference period

Overall, these variations in the number of workers and apprentices have been insufficient to meet labour demand in the construction industry: the job vacancy rates for the building trades have rapidly increased in all the provinces since the COVID-19 pandemic, a trend that started even sooner in Quebec (see Chart 7 below). While labour shortages affect all sectors, the job vacancy rates rose more in the construction sector than in the rest of the economy.Note

Despite the rapid increase in the job vacancy rate, the wages offered to construction workers witnessed a smaller increase between 2015 and 2022 than wages in the rest of the economy, except in Nova Scotia.Note Wages in the construction sector may be slower to adjust to vacancy rates because, in several provinces, these are set by collective agreements that are only periodically renegotiated.Note

Chart 7

Data table for Chart 7 
Data table for Chart 7
Table summary
This table displays the results of Data table for Chart 7. The information is grouped by Province (appearing as row headers), Average hourly wage and Job vacancy rate, calculated using index units of measure (appearing as column headers).
Province Average hourly wage Job vacancy rate
index
Alberta
2015 100.0 100.0
2016 100.7 57.7
2017 100.4 80.0
2018 103.5 89.4
2019 103.1 72.1
2020 110.1 77.9
2021 105.3 142.1
2022 109.1 190.4
British Columbia
2015 100.0 100.0
2016 100.9 116.7
2017 102.7 144.4
2018 108.5 185.4
2019 114.3 150.0
2020 117.6 158.3
2021 126.3 217.4
2022 133.5 254.2
Nova Scotia
2015 100.0 100.0
2016 106.3 73.1
2017 102.3 77.9
2018 101.1 91.3
2019 107.4 103.8
2020 124.9 100.0
2021 113.0 195.2
2022 125.8 255.8
Ontario
2015 100.0 100.0
2016 125.1 120.0
2017 111.4 114.0
2018 109.0 132.0
2019 109.6 136.0
2020 117.8 124.0
2021 121.3 206.0
2022 130.3 247.0
Quebec
2015 100.0 100.0
2016 99.4 75.0
2017 101.4 146.2
2018 103.3 223.1
2019 114.3 280.8
2020 122.6 288.5
2021 124.1 442.3
2022 129.2 453.8

4 Conclusion

The observed evolution of housing prices is the outcome of changes in both the demand for and the supply of housing. Thus, the raw relationships between these costs and each of the indicators discussed in this article should be interpreted as correlations, not as causal relationships.

To illustrate why simple causal analysis is not advised, it can be seen that there is a positive (but weak) correlation between the evolution of prices and the number of new dwelling units approved in building permits. However, this finding does not necessarily imply that more construction causes the price to rise. Rather, it reflects that although increasing housing supply should (all else equal) slow down price growth, new projects are generally launched where there is demand for them. In other words, property owners and builders generally apply for (and municipalities subsequently approve) building projects primarily when the demand for housing is already high. This may induce a positive correlation between house prices and the number of new authorizations. 

To infer causal relationships, the indicators presented in this article should feed more elaborate econometric or quantitative general equilibrium models of local housing markets, keeping in mind that the main drivers of the changes in housing prices may change across cities and time periods (Saiz, 2010; Paciorek, 2013; Accetturo et al., 2021).

The suite of indicators highlighted in this article provides anyone interested in the analysis of housing supply a toolkit with which to describe local conditions relating to the housing supply process. It is hoped that this will facilitate ongoing research in housing affordability.

References

Accetturo, A, Lamorgese, A. R., Mocetti, S., & Pellegrino, D. (2021). Housing supply elasticity and growth: evidence from Italian cities. Journal of Economic Geography, 21(3), 367-396.

Bank of Canada, 2022. Real estate market: Definitions, graphs and data. https://www.bankofcanada.ca/rates/indicators/capacity-and-inflation-pressures/real-estate-market-definitions/

Brooks, L., and Lutz, B., 2016. "From Today's City to Tomorrow's City: An Empirical Investigation of Urban Land Assembly." American Economic Journal: Economic Policy, 8(3), 69-105.

Canada Mortgage and Housing Corporation. (2022). Spring 2022 Housing Market Outlook: Canada and Metropolitan Areas.

Fontaine, J., & Gordon, J., (2023). Residential real estate investors and investment properties in 2020. Housing Statistics in Canada. Statistics Canada Catalogue no. 46280001.

Hilber, C., & Schöni, O., 2022. Housing Policy and Affordable Housing. In LSE Centre for Economic Performance: Occasional Papers 56.

Meyer-Robinson, R., (2022). Prices through the supply chain: Softwood lumber. Prices Analytical Series. Statistics Canada Catalogue no. 62F0014M.

Morissette, R., (2022). Employer responses to labour shortages. Economic and Social Reports 2(7).  Statistics Canada. Statistics Canada Catalogue no. 36-28-0001.

Murphy, A., (2018). A dynamic model of housing supply. American Economic Journal: Economic Policy 104(4), 243-267.

Murray, C. K., (2020).  Time is money: How landbanking constrains housing supply. Journal of Housing Economics 49. https://osf.io/hym43/

Paciorek, A. (2013).  Supply constraints and housing market dynamics. Journal of Urban Economics 77, 11-26.

Saiz, A. (2010). The Geographic Determinants of Housing Supply. The Quarterly Journal of Economics 125(3), 1253-1296.


Appendix A: Details on Data Sources
Table summary
This table displays the results of Appendix A: Details on Data Sources. The information is grouped by Name (appearing as row headers), Link, Geographies, Frequency and First Availability (appearing as column headers).
Name Link Geographies Frequency First Availability
Building construction price index (BCPI) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810027601 CMA Quarterly Q1 1981
Industrial product price index (IPPI) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810026601 Canada Monthly January 1956
New condominium apartment price index (NCAPI) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810027301 CMA Quarterly Q1 2017
New housing price index (NHPI) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1810020501 Canada, geographical region, province/ territory, CMA, Census agglomeration Monthly January 1981
Teranet-National Bank house price index (TN-HPI) https://housepriceindex.ca 11 CMAs Monthly June 1990
Rental market survey (RMS) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3410013301 CMA, Census agglomeration, and Census subdivision Yearly 1987
Investment in building construction (IBC) https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3410017501 Canada, province/ territory, CMA, Census agglomeration Monthly January 2010
Building permits https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3410006601 Canada, province/ territory, CMA, Census agglomeration Monthly January 2011
Housing starts https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3410015601 Canada, province/ territory, CMA Monthly January 1990
Apprenticeships https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3710021901 Canada, geographical region, province/ territory Yearly 1991
Employment https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410002301 Canada, province/territory Yearly 1976
Job vacancy rate, and average offered hourly wage https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1410032601 Canada, province/ territory Quarterly Q1 2015

Date modified: