Analysis in Brief
Short-term rentals in the Canadian housing market

Acknowledgments

The authors would like to thank Dominic Roy and Rowen Stevens for their contributions to this project.

Introduction

The role of short-term rentals (STRs) in Canada’s housing challenges remains a subject of ongoing policy debate in many Canadian cities. While there is a widespread notion that such rentals limit the availability of long-term housing, empirical analysis of their impacts has produced mixed results. This paper provides an overview of STR activity across Canada.

The paper focuses on the subset of STRs that could potentially serve as long-term housing. This subset of STRs, referred to as potential long-term dwellings (PLTDs), is intended to capture STR units that are not serving as anyone’s primary residence, but could potentially function as long-term housing (either as owner-occupied or rental units). The PLTD subset comprises entire units listed for more than 180 days a year, excluding vacation-type properties.

Previous research indicates that STR activity plays an increasingly significant role in the Canadian accommodation services subsector, with its share of revenues rising from an estimated 7.0% in 2017 to 15.2% in 2021.^Note  However, in the housing market, STRs still account for a small proportion of total housing units. In 2023, the estimated number of PLTDs in Canada was 107,266, a figure that represents less than 1% of total housing units in Canada. PLTDs also accounted for a small share of total housing units in Canada’s largest census metropolitan areas (CMAs). However, the share of PLTDs was higher in tourist areas, particularly around ski hills. In Whistler, they constituted 35.0% of all housing units, while in Mont-Tremblant, their share was 16.4%.^Note 

Data and methods

STRs are generally defined as full or partial units made available for rent via online platforms in short-term intervals, typically between 1 and 28 days. These short timelines mean these rentals compete with traditional commercial accommodations for travellers, rather than the renters and lessors of real estate, which operate on monthly and yearly intervals.

The data on Canadian STRs were obtained from AirDNA. AirDNA collects information on STR units listed on Airbnb and Vrbo, the two largest platforms in the STR market.^Note  While information is primarily collected using web scraping, AirDNA also incorporates partner data from channel managers, property managers and individual hosts.^Note 

Housing data used for this analysis come from Statistics Canada and are either directly from the 2021 Census of Population or from the estimates for the stock of occupied private dwellings^Note  produced for intercensal years.^Note  Estimates for the intercensal years are produced at the national and provincial and territorial levels, allowing for a comparison of the PLTD and dwelling counts for these geographic levels for all years up to 2023. However, subprovincial data are available only through the census and, as such, subprovincial analysis was possible only for the 2021 reference year.

The key focus of this paper’s analysis was estimating the number of STR units that could potentially fulfill long-term housing demand if they were not operated for STR purposes. This assessment is important for understanding STR activity.

There are many reasons why a property or dwelling may be rented as an STR unit but would never enter the long-term housing market—for instance, a secondary vacation property rented while the owner resides in their primary residence. Another example is STR listings made for hotel rooms, three-season cottages, boats and other units that are not suitable for long-term housing. Additionally, there are also STR units that primarily serve as long-term housing and thus would not add to the housing supply if they were unlisted. Examples include individual rooms within a residence; student housing leased long term during the academic year and as an STR unit during the summer; and units listed as STRs in the winter by “snowbirds,” who travel south during these months. In the cases outlined above, the STR unit is not depleting the pool of long-term dwellings. Instead, it represents new rental activity that would not have otherwise occurred.

The focus of this analysis is on a specific subset of STRs—PLTDs. As mentioned in the introduction, PLTDs comprise entire units listed for more than 180 days a year, excluding vacation-type properties (for more information on the definition of PLTDs, see Appendix A). The full breakdown of listing types for 2023 shows that PLTDs accounted for 30.2% of listings, while roughly one-third (33.5%) of listings were entire units that were unavailable for the majority of the year, 15.0% were entire unit listings for vacation-type properties and the remaining 21.3% of listings were partial units (Chart 1). For the various reasons considered above, none of the final three categories are considered to be units that could satisfy long-term demand.

Data table for Chart 1

Data table for Chart 1 Table summary
This table displays the results of Data table for Chart 1 Short-term rentals and Shares , calculated using number of units and percent units of measure (appearing as column headers).

	Short-term rentals	Shares
	number of units	percent
Source: Statistics Canada, custom tabulation from AirDNA data.
Potential long-term dwellings	107,266	30.2
Entire unit—unavailable majority of year	118,934	33.5
Entire unit—vacation types	53,407	15.0
Partial units (any type)	75,462	21.3

The PLTD estimates are obtained using data from AirDNA. No surveys were conducted, and no ownership data can be linked to specific STR units to better understand which units could feasibly be used as long-term dwellings. Caution should be used when interpreting the PLTD figures, and they should be considered only an estimate, not an actual count of dwellings that have been removed from the long-term housing stock. Nevertheless, the PLTD estimates provide a more reliable indicator than assuming all Canadian STR listings, over 355,000 in 2023, have the capacity to function as long-term dwellings.

Current literature

Several studies have investigated the influence of private STRs on rental markets and housing shortages. The findings from these studies offer insights into the extent to which STR activity affects the housing market.

A recent Conference Board of Canada report^Note  suggests that the level of Airbnb activity had no meaningful impact on the cost of rent, stating that “the share of dwellings used for Airbnb activity is too small in most neighbourhoods—on average less than 0.5 per cent—to have a meaningful impact.” That estimate was determined based on what it referred to as “high-use Airbnb” units, defined as “an entire home or apartment that has been rented out for more than 30 nights in the previous three months, and likely to be a full-time short-term rental and therefore unlikely to be a host’s principal place of residence.”

McGill University’s Urban Politics and Governance research group has also published several papers examining the impact of STR activity on housing availability. In a 2017 paper,^Note  researchers associated with the group attempted to estimate the number of units removed from the housing supply because of STR activity in Canada’s three largest cities. They found that “there are now 13,700 entire homes rented 60 days or more per year on Airbnb in Montreal, Toronto and Vancouver, each of which is unlikely to be rented to long-term tenants. These entire homes account for one sixth of all Airbnb listings, and a majority of nights booked on the service.” The number comes from their calculation of “full-time Airbnb” use. They defined this concept as the number of days per year that a unit is booked (“occupancy”) and the number of days that a unit is either booked or available to be booked (“availability”). They define “full-time” as 60 days of occupancy and 120 days of availability.

In 2019, members of the group published another paper,^Note  which used the concept of frequently rented entire-home (FREH) listings to identify the subset of STRs that may be removing units from the housing supply. They define FREH listings as STRs that were “available for rent at least half the year (183 nights) and actually rented at least 90 nights. FREH listings represent a conservative estimate for housing either directly converted to STRs or under serious threat of conversion since it is highly unlikely that a home that spends the majority of the year listed on Airbnb is housing a long-term resident.” This estimate suggests that Airbnb has removed approximately 31,100 units of housing from the long-term rental market.

These studies show that there is no standardized measurement for estimating the number of dwellings being removed from the long-term housing market because of STR activity. The FREH concept aligns more closely with the PLTD concept defined in this paper. A notable difference with all three referenced papers is the use of a threshold for days rented. The PLTD concept in this analysis did not account for days rented, since whether a unit was successfully rented out is not a requirement for removing it from the long-term stock. An empty unit listed on an STR platform could otherwise accommodate a long-term tenant or owner but currently remains vacant.

Another distinction between PLTDs and the concepts used in these papers is the threshold used for the number of days listed. The Conference Board of Canada paper, for example, employs a shorter listing time threshold of 120 days. Shorter timelines may capture units that are made available for only part of the year and still primarily function as long-term housing, such as units listed during the winter months by snowbirds.

A final difference between PLTDs and these other concepts is the exclusion of certain property types. Specifically, the PLTD estimates try to exclude vacation-type properties such as cottages, purpose-built vacation homes and other vacation properties that would be unlikely to enter the long-term housing market.^Note 

In summary, these studies show previous efforts made to understand STR activity in Canada. The present study reinforces the use of the PLTD concept in an environment with no standardized measurement practices.

Results

Recent trends for short-term rentals

Total STR listings increased by more than 60% in Canada from 2017 to 2023, while the number of PLTDs grew by more than 80%, from 58,441 to 107,266 units, over the same period (Chart 2). As a result, the share of listings considered as PLTDs rose from 27.2% of total listings in 2017 to 30.2% in 2023 (Chart 3).

Data table for Chart 2

Data table for Chart 2 Table summary
This table displays the results of Data table for Chart 2 All short-term rentals and Potential long-term dwellings, calculated using number of units units of measure (appearing as column headers).

	All short-term rentals	Potential long-term dwellings
	number of units
Source: Statistics Canada, custom tabulation from AirDNA data.
2017	214,808	58,441
2018	267,634	74,083
2019	303,521	88,494
2020	266,444	72,796
2021	245,109	63,589
2022	278,841	70,139
2023	355,070	107,266

Yet the progression did not follow a linear path, with STR activity declining after the onset of the COVID-19 pandemic. Total listings in Canada fell 19.2% from 2019 to 2021, while the PLTD subset decreased even more, by 28.1% over the same two years. In 2022, even as STR activity started to pick up, the share of PLTDs (25.2% of total listings) was still lower than in 2019 (29.2%).

Data table for Chart 3

Data table for Chart 3 Table summary
This table displays the results of . The information is grouped by (appearing as row headers), , calculated using (appearing as column headers).

	Percent
Source: Statistics Canada, custom tabulation from AirDNA data.
2017	27.2
2018	27.7
2019	29.2
2020	27.3
2021	25.9
2022	25.2
2023	30.2

As mentioned previously, the decline in PLTDs during the pandemic was more severe than the overall decrease in STR activity. This greater decline in the PLTD subset may support the notion that these units could be used as long-term dwellings. After the decline in tourism during the pandemic, many property owners may have converted their STRs to long-term rentals. This could also explain why 2022 marked a low point in the proportion of PLTDs, since many thousands of units may have still been tied up in 12-month leases during the onset of the recovery. However, this assumption could be confirmed only with property ownership data, which are not available for this analysis.

Short-term rentals and total housing units

Housing stock data^Note  for the intercensal years indicate that there were 15.5 million housing units^Note  in Canada in the last quarter of 2023. This highlights a pronounced disparity in scale, with the total number of dwellings being orders of magnitude larger than the estimate of PLTDs. At the national level, PLTDs accounted for 0.69% of Canadian housing units in 2023 (Chart 4). This figure is an all-time high for Canada, with the previous high of 0.60% occurring in 2019.

Data table for Chart 4

Data table for Chart 4 Table summary
This table displays the results of . The information is grouped by (appearing as row headers), , calculated using (appearing as column headers).

	Percent
Sources: Statistics Canada, Table 36-10-0688-01 and custom tabulation from AirDNA data.
2017	0.41
2018	0.51
2019	0.60
2020	0.49
2021	0.42
2022	0.46
2023	0.69

These trends differ at the provincial level.^Note  In Ontario, the share of housing units defined as PLTDs more than doubled, jumping from 0.35% in 2022 to an all-time high of 0.69% in 2023 (Table 1). In Quebec, there was a jump from 0.38% in 2022 to 0.51% in 2023. However, this did not exceed Quebec’s pre-pandemic high of 0.61%, which occurred in 2019. It is possible that these differences are the result of different regulatory approaches, since Quebec has enacted province-wide STR regulations, while regulations have been enacted only at the municipal level in Ontario.

At the provincial level, only British Columbia and Prince Edward Island had a share of PLTDs that exceeded 1% of housing units in 2023. This finding aligns with those provinces being the leaders in STRs, with their STR markets claiming the greatest share of revenue within their respective accommodation services subsectors.^Note 

Housing unit estimates are not available at the subprovincial level between census years. As a result, the following estimates are available only for 2021. In 2021, PLTDs accounted for less than half a percent of housing units in Canada’s five largest CMAs by population (Table 2). Additionally, among the largest CMAs, only Vancouver (0.45%) had a PLTD share of housing units that exceeded the 2021 national average of 0.42%. These findings are similar to those of the Conference Board of Canada report mentioned in the “Current literature” section, which showed that “on average less than 0.5 per cent” of dwellings were high-use Airbnb units in the neighbourhoods it studied.^Note 

The shares were higher in tourist areas, especially in ski towns. Whistler had the highest share by far in 2021, with 35.0% of housing units being PLTDs (Table 3). A situation in which PLTDs make up more than one-third of housing units can be expected to have a significant impact on a community’s housing market. However, the nature of the market as a tourist hotspot likely changes the approach to STRs for policy makers and other stakeholders. These areas may be disproportionately reliant on STR activity since it often supports tourism and stimulates the local economy. Other popular tourist markets in more rural areas, such as Mont-Tremblant (16.4%), Canmore (15.0%) and The Blue Mountains (13.2%), all have similar shares of PLTDs as part of their housing supply. The Prince Edward County census subdivision had the fifth-highest share of PLTDs, at 4.9%.

Concluding thoughts

This analysis has shown that the subset of STR units capable of serving as long-term housing, defined as PLTDs, is generally small in most Canadian markets. The degree to which STR activity impacts housing affordability was not a focus of this paper, and so the results should not necessarily be used to draw conclusions on price impacts without further analysis.

Housing market dynamics are complex,^Note  and there is unlikely to be a simple and straightforward solution to the current challenges of affordability and supply faced by many Canadians. This paper has focused on STR activity within the housing market. However, it is important to acknowledge the influence of many other factors affecting affordability and supply, including, but not limited to, multiple-property owner investors,^Note  the housing supply in relation to population growth,^Note  and factors relating to interest rates and financing.

Responding to concerns regarding STR activity, numerous municipalities^Note Note  and some provinces^Note Note  have enacted or strengthened regulations. Additionally, in its 2023 Fall Economic Statement, the federal government introduced new tax policies targeting non-compliant STR operators.^Note  This analysis offers a clearer understanding of STR activity across Canada and its relation to the Canadian housing market. For detailed data from this analysis, refer to the appendices.

Appendix A: Definition of potential long-term dwellings

The term “potential long-term dwellings” in this paper refers to the subset of Canadian short-term rental units that satisfy the following conditions:

1. The listing on Airbnb and/or Vrbo is for an entire unit.
2. The unit is listed for at least 180 days a year.
3. The property type description provided by the Airbnb or Vrbo host does not correspond to the list of vacation-type properties outlined below.

Appendix B: Short-term rental and potential long-term dwelling data, Canada, provinces and territories, 2017 to 2023

Appendix C: Short-term rental and potential long-term dwelling data, census metropolitan areas, 2021

Appendix D: Short-term rental and potential long-term dwelling data, census subdivisions with at least 50 potential long-term dwellings, 2021