Housing Statistics in Canada
Measuring unmet housing need and housing instability in households with roommates and extended family
By Aaron Gorski, David Heisel and Jerry Situ
Skip to text
Text begins
Highlights
- In 2021, 1.65 million households comprised of roommates or extended family sharing living space. These households were about evenly split between those with roommates and those with extended family, each group representing about 800,000 households.
- One in five households with roommates or extended family members (21.7%) was living in a crowded dwelling, compared with 3.4% of other households. In contrast, households with roommates or extended family members (16.7%) were less likely to be in unaffordable housing than other households (21.5%).
- About 900,000 people lived with extended family without contributing to housing costs, and almost half (47.5%) of them had no income or an income of less than $30,000.
- Just under 400,000 people lived with non-relatives without contributing to housing costs, and over one-third (36.6%) of them were living in poverty.
- Estimates of the number of people in shared housing experiencing housing instability or unmet housing need varied depending on the criteria used to define these concepts. One estimate suggested that 71,000 people had several risk factors for housing instability or unmet housing need, including living with non-relatives, not contributing to housing costs, having an income of less than $30,000 or living in poverty, and residing in a crowded dwelling. Another estimate indicated that just under 1.7 million people could be experiencing housing instability or unmet housing need when defined solely by living in a crowded dwelling.
Introduction
Canada is facing housing affordability challenges. In 2021, one in five households (20.9%) lived in unaffordable housing, defined as spending 30% or more of household total income on shelter costs (Statistics Canada, 2022c). Some estimates have projected a need for an additional 3.5 million housing units by 2030 to restore housing affordability (Canada Mortgage and Housing Corporation, 2022).
Living with roommates or with extended family members is one strategy that may be used to manage housing affordability challenges. Roommate households, defined as two or more people who do not form a census family living together, represented a small share (4%) of Canada’s households in 2021. Nevertheless, it was the fastest-growing household type, with the number of roommate households increasing by 54% from 2001 to 2021. Moreover, multigenerational households, multiple-census-family households, and households with one census family plus additional people have also grown relatively rapidly from 2001 to 2021 (45%) and represented 7% (or 986,400) of all households in 2021 (Statistics Canada, 2022b).
Although living with others can improve affordability for the household, individual members may still be at risk of losing their housing (referred to as “housing instability” in this paper) or may not have the housing that meets their individual needs (referred to as “unmet housing need” in this paper).
Examining these households can reveal sources of housing instability risk and unmet housing need that are not apparent in household count projections or affordability statistics based on the combined resources of all household members. Studies in the United States have looked at changes in the level of “doubled-up” households—households with one or more adults in addition to the head of the household and their spouse or partner—as an indicator of housing vulnerability (Housing and Urban Development, n.d.). Similarly, studies in the United Kingdom and European countries have looked at “concealed” households—family units or single adults living within “host” households—in relation to housing need, hidden homelessness and the impact on crowding (Bramley et al., 2010; Office for National Statistics, 2023).
Not all households with roommates or extended family members are at risk of housing instability or unmet housing need, as reasons for cohabiting can be social or cultural, not just economic. International measures, like doubled-up or concealed households, often factor in other characteristics, such as income, school enrolment, age and crowding. Other studies have applied additional exclusions with the aim of focusing on households at a higher risk of housing instability or unmet housing need. The definitions vary across studies, often based on the analytical objective and available data (Richard et al., 2022).
This paper uses data from the 2021 Census of Population to provide a nuanced look at households with roommates or extended family members. First, it describes living arrangements and how housing costs are shared among household members within these households. Second, it examines the housing and socioeconomic characteristics of individuals in these house-sharing and cost-sharing arrangements to illustrate varying levels of housing instability risks and risks of unmet housing need. Lastly, it assesses some risk profiles to estimate the size of the population in these households that could be at risk of housing instability or unmet housing need.
This analysis describes the characteristics of households with roommates and extended family to inform the discussion on how best to measure and track housing instability and unmet housing need within these households.
Methodology
The approach used in this paper borrows from the research on doubled-up households in the United States (Richard et al., 2022). Specifically, it examines non-nuclear-family households and uses additional characteristics to identify the subgroups at higher risk of housing instability or unmet housing need.
This analysis uses data from the 2021 Census of Population. The census is a suitable data source because it collects detailed relationship information and housing characteristics, which are necessary to classify doubled-up households. However, as the census enumerates people based on their usual place of residence,Note roommate living arrangements where household members are enumerated elsewhere are not captured in the census. Examples of these roommate households are student roommates who are enumerated at their parental home or workers who share housing but return to their usual home elsewhere.
The definition of doubled-up households does not focus on a specific type of housing instability or unmet housing need. In fact, there is no standard measure of housing instability (Frederick et al., 2014). Similarly, unmet housing need can have various meanings and be defined based on many dimensions, such as crowding, adequacy and affordability. While the specific level and type of housing instability and unmet housing need can be different under different definitions of doubled-up households, measuring them over time helps track changes and trends in housing instability and unmet housing need.
The population analyzed in this paper is households with roommates or extended family members. These households are similar to the non-nuclear-family households used in international measures. The term “house sharing” is used in this paper to represent the non-nuclear-family living arrangement and is similar to the concepts of doubled-up or concealed households. This paper defines house sharing as a private household with two or more people who are not in the same census family (these are generally roommate households or households with extended family).Note In the 2021 Census, house-sharing households represent 1.65 million out of 15.0 million private households. Other households, which include one-person households and households with one census family only, are not house sharing and are excluded from this analysis.
The characteristics explored in this paper were based on key attributes found in the literature that were important for defining doubled-up households and were available in the census data. They include
- household living arrangements
- financial contribution towards housing costs from household members
- housing need indicators, such as crowding and core housing need
- economic hardship indicators, such as low income, poverty and affordability.
Appendix A provides the definition of each of these variables and the background on why they were included in this analysis.
The 1.65 million households with roommates or extended family were about evenly split between those with non-relatives and those with relatives only
In 2021, there were 1.65 million households with roommates or extended family members (referred to as house sharing). Among these households, the distribution was about evenly split between households with relatives only and those with non-relatives (each accounting for about 800,000 households). While the number of households was about the same, relatives-only households and non-relatives householdsNote were different types of households, as shown in Table 1.
For house-sharing situations among relatives-only households, many of them were multigenerational households (53.3% for one-payer households and 43.3% for multi-payer households).Note While multigenerational living can be motivated by housing costs, it can be done for other reasons, such as care, family reunification, or cultural or individual preferences (Statistics Canada, 2022b).
For house-sharing among non-relatives, most households (54.3% for one-payer households and 70.3% for multi-payer households) consisted of people not in a census family, such as roommate situations. This group, particularly those who also shared housing costs, predominantly rented their dwelling. Renting a place with roommates is a common step in the housing ladder, often associated with a period of transition or financial constraint.
The way housing costs were shared among household members also differed between relatives-only households and non-relatives households. Sharing housing costs among household members was more common in households with non-relatives than in those with relatives only. Among relatives-only households, one-payer households outnumbered multi-payer households, while the opposite was true among non-relatives households. This difference indicates that cost-sharing arrangements should be considered with house-sharing arrangements. When house sharing with relatives, familial ties can mitigate the risks that come with not contributing to housing costs.
Categorizing households by their house-sharing and cost-sharing arrangements can reveal different reasons for cohabiting and, consequently, different risks of housing instability or unmet housing need. House sharing with relatives was associated with cohabiting arrangements that are often motivated by non-economic reasons (e.g., multigenerational living), whereas house sharing with non-relatives was associated with cohabiting arrangements often motivated by costs (e.g., renting with roommates).
| Household types | Household group based on living arrangement and maintainer status | ||||
|---|---|---|---|---|---|
| Relatives only, multi-payer | Relatives only, one payer | Non-relatives, multi-payer | Non-relatives, one payer | Total | |
| number of households | |||||
| Source: Statistics Canada, Census of Population, 2021. | |||||
| Multigenerational household | 159,880 | 265,420 | 12,760 | 7,640 | 445,700 |
| Other census family household: Multiple-census-family household | 30,040 | 36,585 | 17,625 | 6,550 | 90,800 |
| Other census family household: One-census-family household with additional people | 84,495 | 139,300 | 134,335 | 91,940 | 450,070 |
| Two-or-more-people, non-census-family household | 94,405 | 56,245 | 390,600 | 125,940 | 667,190 |
| Total | 368,825 | 497,550 | 555,325 | 232,065 | 1,653,760 |
| percent of households | |||||
| Multigenerational household | 43.3 | 53.3 | 2.3 | 3.3 | 27.0 |
| Other census family household: Multiple-census-family household | 8.1 | 7.4 | 3.2 | 2.8 | 5.5 |
| Other census family household: One-census-family household with additional people | 22.9 | 28.0 | 24.2 | 39.6 | 27.2 |
| Two-or-more-people, non-census-family household | 25.6 | 11.3 | 70.3 | 54.3 | 40.3 |
| Total | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
One in five households with roommates or extended family were living in crowded dwellings
The previous section demonstrated that distinguishing roommate households from extended family households based on their house-sharing and cost-sharing arrangements can categorize them into groups with potentially different risk levels of housing instability and unmet housing need. This section will look at the housing characteristics of these groups to determine whether there are indeed indications of higher levels of risk.
Across the different types of house-sharing situations, about one in five households (21.7%) was living in a crowded dwelling, compared with 3.4% of households that do not house share (Table 2).Note Crowding is a common variable used internationally to identify situations that reflect instability or unmet housing need.Note
The proportion of households in dwellings requiring major repairs was also higher for house-sharing households (7.5%) than for non-house-sharing households (6.0%). Crowded dwellings often experience a higher incidence of requiring major repairs than uncrowded dwellings (Lorentzen et al., 2022).
While the incidence of crowding and the need for major repairs was higher among house-sharing households, the incidence of unaffordable housing was lower (16.7% for house-sharing households versus 21.5% for non-house-sharing households). However, because the housing affordability indicator is based on the household’s total income, it does not reflect housing affordability for individual household members. Moreover, because many house-sharing households have members who do not contribute to housing costs, household total income may not accurately reflect housing affordability for the members who are responsible for housing costs.
The level of core housing need was also lower among house-sharing households (6.8%) than among non-house-sharing ones (10.5%). Like the affordability indicator, core housing need is calculated based on the combined income of all household members.
The lower incidence of unaffordable housing and core housing need among house-sharing households reinforces that contributions to housing costs can be an important factor in differentiating housing risks. While the affordability and core need indicators may not accurately reflect the situation when housing costs are not shared, the lower incidence of unaffordability and core need when costs are shared is evidence of lower risks.
In summary, house-sharing households have higher risks of housing instability or unmet housing need than non-house-sharing households, as indicated by the higher prevalence of crowding and the need for major repairs. However, other indicators of housing risk or unmet housing need, including housing affordability and core need, were not higher and may point to a lower risk of housing instability or unmet housing need (especially when housing costs are shared among household members).
| Household characteristics | House sharing: Household group based on living arrangement and maintainer status | Not house sharing | ||||
|---|---|---|---|---|---|---|
| Relatives only, multi-payer | Relatives only, one payer | Non-relatives, multi-payer | Non-relatives, one payer | Total | ||
| number of households | ||||||
|
||||||
| All households | 368,825 | 497,550 | 555,325 | 232,065 | 1,653,760 | 13,325,180 |
| percent of households | ||||||
| Shelter-cost-to-income ratio Table 2 Household characteristics by house-sharing arrangement, living arrangement and maintainer status, Canada, 2021 Note 1 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Less than 30% | 85.2 | 86.9 | 78.6 | 84.5 | 83.3 | 78.5 |
| 30% or more | 14.8 | 13.1 | 21.4 | 15.5 | 16.7 | 21.5 |
| Crowding | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Not crowded | 77.4 | 79.0 | 77.6 | 79.7 | 78.3 | 96.6 |
| Crowded | 22.6 | 21.0 | 22.4 | 20.3 | 21.7 | 3.4 |
| Condition of dwelling | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Regular maintenance or minor repairs needed | 92.2 | 93.0 | 92.4 | 92.5 | 92.5 | 94.0 |
| Major repairs needed | 7.8 | 7.0 | 7.6 | 7.5 | 7.5 | 6.0 |
| Core housing need Table 2 Household characteristics by house-sharing arrangement, living arrangement and maintainer status, Canada, 2021 Note 2 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Not in core need | 93.7 | 94.3 | 92.3 | 92.2 | 93.2 | 89.5 |
| In core need | 6.3 | 5.7 | 7.7 | 7.8 | 6.8 | 10.5 |
| Tenure Table 2 Household characteristics by house-sharing arrangement, living arrangement and maintainer status, Canada, 2021 Note 3 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Owner | 64.8 | 82.9 | 33.7 | 67.6 | 60.0 | 67.6 |
| Renter | 35.2 | 17.1 | 66.3 | 32.4 | 40.0 | 32.4 |
| Household size | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| 1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 33.0 |
| 2 | 22.1 | 9.7 | 51.2 | 44.5 | 31.3 | 34.6 |
| 3 | 20.1 | 19.2 | 23.1 | 21.9 | 21.1 | 13.9 |
| 4 | 18.5 | 22.4 | 12.1 | 14.2 | 16.9 | 12.9 |
| 5 | 16.9 | 23.6 | 6.5 | 9.9 | 14.4 | 4.2 |
| 6 | 11.8 | 15.3 | 3.5 | 5.3 | 9.1 | 1.1 |
| 7 | 5.6 | 5.9 | 1.8 | 2.3 | 3.9 | 0.3 |
| 8 or more | 5.1 | 3.8 | 1.9 | 2.0 | 3.2 | 0.1 |
Just under 400,000 people lived with non-relatives without contributing to housing costs, and over one-third (36.6%) of them were living in poverty
The previous section showed that households with different house- and cost-sharing arrangements can have different risks of housing instability and unmet housing need based on common housing indicators. This section looks at the characteristics of the people in house-sharing situations to understand their individual risks and their ability to cope with them.
Table 3 shows the characteristics of people in house-sharing situations with relatives only and non-relatives, categorized by whether they or their census family contributed to housing costs.Note This distinction is made because of the significant role contributions to housing costs can play in the risk of unmet housing need and housing instability.
| Characteristics | House sharing among related people | Housing sharing among unrelated people | Total house sharing | Not housing sharing | ||
|---|---|---|---|---|---|---|
| Person or family contributes to housing costs Table 3 Characteristics of people by house-sharing arrangement, relationship and contribution to housing costs, Canada, 2021 Note 4 | Neither person nor family contributes to housing costs | Person or family contributes to housing costs Table 3 Characteristics of people by house-sharing arrangement, relationship and contribution to housing costs, Canada, 2021 Note 4 | Neither person nor family contributes to housing costs | |||
| number of people | ||||||
|
||||||
| All people | 2,891,190 | 899,695 | 2,049,915 | 388,415 | 6,229,220 | 30,099,260 |
| percent of population | ||||||
| Gender Table 3 Characteristics of people by house-sharing arrangement, relationship and contribution to housing costs, Canada, 2021 Note 1 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Women+ Table 3 Characteristics of people by house-sharing arrangement, relationship and contribution to housing costs, Canada, 2021 Note 2 | 51.7 | 56.1 | 45.6 | 46.5 | 50.0 | 50.7 |
| Men+ Table 3 Characteristics of people by house-sharing arrangement, relationship and contribution to housing costs, Canada, 2021 Note 3 | 48.3 | 43.9 | 54.4 | 53.5 | 50.0 | 49.3 |
| Age | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Younger than 15 years | 18.7 | 19.3 | 8.3 | 4.4 | 14.5 | 16.9 |
| 15 to 24 years | 12.9 | 13.5 | 21.2 | 22.7 | 16.3 | 10.5 |
| 25 to 34 years | 14.5 | 15.3 | 27.9 | 24.3 | 19.7 | 12.2 |
| 35 to 44 years | 13.7 | 8.2 | 12.7 | 12.8 | 12.5 | 13.6 |
| 45 to 54 years | 13.5 | 5.0 | 10.9 | 11.0 | 11.3 | 13.1 |
| 55 to 64 years | 13.8 | 8.6 | 10.6 | 12.2 | 11.9 | 14.7 |
| 65 to 74 years | 8.5 | 14.1 | 5.9 | 8.1 | 8.4 | 11.5 |
| 75 to 84 years | 3.3 | 10.8 | 2.0 | 3.4 | 4.0 | 5.7 |
| 85 years and older | 1.1 | 5.2 | 0.5 | 1.0 | 1.5 | 1.8 |
| School attendance for people 15 years and older | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Did not attend school | 85.7 | 86.9 | 79.4 | 81.2 | 83.3 | 87.0 |
| Attended elementary or secondary school only | 5.1 | 5.0 | 2.7 | 4.9 | 4.2 | 4.6 |
| Attended postsecondary school | 9.2 | 8.1 | 17.9 | 13.8 | 12.4 | 8.3 |
| Labour force status for people 15 years and older | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| In the labour force, employed | 58.2 | 33.3 | 64.5 | 49.2 | 56.4 | 57.2 |
| In the labour force, unemployed | 7.6 | 6.9 | 9.3 | 9.3 | 8.2 | 6.2 |
| Not in the labour force | 34.1 | 59.8 | 26.2 | 41.5 | 35.3 | 36.6 |
| Individual Market Basket Measure poverty status | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Not applicable (reserves) | 2.5 | 4.7 | 0.7 | 1.7 | 2.2 | 0.8 |
| In poverty | 3.7 | 4.8 | 24.4 | 36.6 | 12.7 | 7.1 |
| Not in poverty | 93.8 | 90.5 | 74.9 | 61.7 | 85.1 | 92.2 |
| Income of census family or person not in a census family | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 | 100.0 |
| Without total income | 0.4 | 9.2 | 1.8 | 8.1 | 2.6 | 0.2 |
| Under $10,000 (including loss) | 1.7 | 10.1 | 5.1 | 11.0 | 4.6 | 0.9 |
| $10,000 to $19,999 | 3.0 | 9.9 | 9.0 | 13.8 | 6.7 | 1.8 |
| $20,000 to $29,999 | 6.3 | 18.3 | 14.7 | 20.4 | 11.7 | 4.4 |
| $30,000 to $39,999 | 6.6 | 12.9 | 12.7 | 14.5 | 10.0 | 5.1 |
| $40,000 to $49,999 | 6.7 | 8.7 | 10.4 | 10.0 | 8.4 | 5.9 |
| $50,000 to $59,999 | 6.9 | 6.8 | 8.4 | 6.8 | 7.4 | 6.2 |
| $60,000 to $69,999 | 6.9 | 5.2 | 6.8 | 4.5 | 6.5 | 6.4 |
| $70,000 to $79,999 | 6.7 | 4.2 | 5.5 | 3.2 | 5.7 | 6.4 |
| $80,000 to $89,999 | 6.4 | 3.3 | 4.5 | 2.2 | 5.1 | 6.3 |
| $90,000 to $99,999 | 6.2 | 2.6 | 3.7 | 1.5 | 4.5 | 6.1 |
| $100,000 to $149,999 | 22.4 | 6.3 | 10.2 | 3.0 | 14.8 | 23.9 |
| $150,000 and over | 19.7 | 2.6 | 7.1 | 1.0 | 11.9 | 26.4 |
There were 6.2 million people living in house-sharing situations. Among them, 60.9% were living with relatives only and 39.1% were living with non-relatives. About 900,000 people did not contribute to housing costs in relatives-only households, and 388,000 people did not contribute in non-relatives households.
In relatives-only households, people who contributed to housing costs generally had a similar characteristic profile as those who did not house share. In particular, the income distribution was similar, and the incidence of poverty was lower. Therefore, no specific characteristics suggest this group is at higher risk of housing instability or unmet housing need.
In contrast, people who did not contribute to housing costs in relatives-only households had different characteristics, compared with those in non-house-sharing households. Specifically, a higher proportion of those not contributing to costs in relatives-only households were older (30.1% were aged 65 or older versus 19.0% for people in non-house-sharing households) and were women (56.1% versus 50.7%). Moreover, in terms of finance-related characteristics, a higher proportion of those not contributing to costs in relatives-only households were out of the labour force (59.8% versus 36.6% for people in non-house-sharing households; population aged 15 or older) and had lower incomeNote (47.5% had no income or an income of less than $30,000 in 2020 versus 7.3% for people in non-house-sharing households).
People in non-relatives households—regardless of whether they were contributing to housing costs—had a different characteristic profile, compared with those who were not house-sharing. A higher proportion of people house sharing were men (54.3% compared with 49.3% for people who did not house share), were aged 15 to 34 (48.8% versus 22.7%), were living in poverty (26.3% versus 7.1%) and had no income or an income of less than $30,000 in 2020 (34.2% versus 7.3%).
Among the people in non-relatives households, those not contributing to housing costs were particularly economically disadvantaged. Over one-third (36.6%) were living in poverty, and 53.3% had no income or an income of less than $30,000. Among those 15 years and older, 41.5% were not in the labour force and 9.3% were unemployed.
Differences in characteristics between the groups emerged when they were categorized based on house- and cost-sharing arrangements. These differences suggest varying risk levels of housing instability and unmet housing need. People in relatives-only households who contribute to housing costs were similar to those in non-house-sharing households, while people in other groups had characteristics indicative of a higher risk of housing instability and unmet housing need. People in non-relatives households who do not contribute to housing costs had characteristics that suggest they had the highest level of risk among the groups.
Estimates of the number of people experiencing housing instability or unmet housing need in shared housing ranged from 71,000 to just under 1.7 million, depending on the criteria used in the definition
The previous sections demonstrated that a higher proportion of people in house-sharing households had certain characteristics related to housing instability and unmet housing need, compared with those in non-house-sharing households. However, no single set of characteristics definitively identifies unmet housing need or determines housing instability. Therefore, this section provides estimates of some risk profiles to inform and stimulate discussion on how to define measures like doubled-up households, which can be used to track trends in housing instability and unmet housing need among house-sharing households.
The risk profiles and the corresponding estimates are shown in Table 4. The first profile to be estimated is the number of people living in crowded dwellings. This profile was considered because in some housing need metrics, such as acceptable housing, crowding indicates unmet housing need. Among the 6.2 million people in house-sharing households, just under 1.7 million were living in crowded dwellings.
The next profile included people with no family income, people with a family income of less than $30,000 or people living in poverty. This profile was considered because in some housing metrics, such as core housing need, income can indicate available alternatives and therefore the ability to cope with housing instability. Just under 1.7 million people were in this scenario.
Not contributing to housing costs in non-relatives households was also a profile, as people in this situation have weaker ties to the rest of the household members, potentially increasing their risk of housing instability. Just under 400,000 people were cohabiting with non-relatives without contributing to housing costs.
Combinations of the above risk profiles were also explored to identify smaller groups with a higher risk of housing instability or unmet housing need. For example, 122,000 people who were cohabiting with non-relatives without contributing to housing costs were also in crowded dwellings. When this group was further limited to those without a family income, with a family income of less than $30,000 or who were living in poverty, the number of individuals fell to 71,000.
Whether or not to include people attending a postsecondary education institution was also a consideration, given they are treated differently in other housing indicators, like core housing need. The estimates for both options are found in Table 4. Excluding those pursuing postsecondary education reduced the estimate by about 12% to 21%, depending on the scenario. In the most restrictive scenario, the estimate is reduced to 56,000 from 71,000.
Estimates for the number of people who were house sharing and at risk of housing instability or unmet housing need varied depending on the definition of the indicator being used. When based solely on low income or crowding, the estimate was as high as 1.7 million people. However, the estimate was as low as 71,000 when it was based on multiple criteria, including low income, crowding, and living with non-relatives without contributing to housing costs.
| Risk profile | Living in house-sharing households (A) |
Living in house-sharing households but not attending postsecondary school (B) |
Living in house-sharing households and attending postsecondary school (A minus B) |
|---|---|---|---|
| number of people | |||
| Source: Statistics Canada, Census of Population, 2021. | |||
| Total population | 6,229,220 | 5,566,685 | 662,535 |
| Lived in an unsuitable dwelling | 1,687,220 | 1,476,545 | 210,675 |
| Had no income, had an income of less than $30,000 or lived in poverty | 1,689,445 | 1,387,910 | 301,535 |
| Did not contribute to housing costs in non-relatives households | 388,415 | 336,985 | 51,430 |
| Did not contribute to housing costs in non-relatives households and lived in an unsuitable dwelling | 121,500 | 101,995 | 19,505 |
| Did not contribute to housing costs in non-relatives households; lived in an unsuitable dwelling; and had no income, had an income of less than $30,000 or lived in poverty | 70,960 | 56,145 | 14,815 |
Conclusion
These findings from the 2021 Census present a nuanced picture of households with roommates or extended family, emphasizing the subgroups that are more likely to experience unmet housing need and housing instability. The data reveal these house-sharing households had a higher incidence of crowding (21.7% versus 3.4% for non-house-sharing households), but also a lower incidence of unaffordable housing (16.7% versus 21.5%) and core housing need (6.8% versus 10.5%).
Disaggregating house-sharing situations based on living arrangements and responsibility for housing costs revealed separate groups with varying risks of unmet housing need and housing instability. Notably, about 900,000 people shared housing with relatives only without paying housing costs, and almost half (47.5%) of them had family income below $30,000. Moreover, just under 400,000 people shared housing with non-relatives without paying housing costs, and over one-third (36.6%) of them were living in poverty.
Various risk profiles were explored to identify a range of estimates of the population in house-sharing households that could be at risk of housing instability or unmet housing need. The number of people with multiple risk factors—including living with non-relatives and not contributing to housing costs, having an income of less than $30,000 or living in poverty, and residing in a crowded dwelling—was estimated at 71,000. In contrast, estimates were higher when looking at just one risk factor. For example, the number of people could be as high as 1.7 million, which is the estimate for people living crowded dwellings.
Although the estimates for housing instability and unmet housing need in house-sharing households can vary depending on which groups are considered at risk, increases in the prevalence of house sharing, as seen recently in Canada, can be an indicator of rising levels of unmet housing need and housing instability. This underscores the importance of considering these groups when studying and addressing housing supply and demand challenges.
Appendix A: Definitions of concepts used in the analysis
This appendix provides the definition of the variables used in the analysis and background on why they were included.
Household living arrangement
Whether a dwelling is shared among related or unrelated people is examined to differentiate households based on the living arrangements of the household members. House sharing with relatives versus non-relatives is an important distinction for understanding the social and cultural reasons for cohabitation. Multigenerational living arrangements play a growing role in care and social support networks (Statistics Canada, 2022b). Likewise, the way economic resources are combined likely varies between relatives and non-relatives.
American studies on doubled-up households have also differentiated between households with grandchildren, other relatives and non-relatives (Richard et al., 2022). Moreover, qualitative research has established important care and parenting dynamics in doubled-up households with relatives (Harvey et al., 2021).
In this paper, house sharing with relatives or non-relatives is based on the definitions of census family and economic family. A census family is a couple, with or without children, or a one-parent family. An economic family includes all people related by blood, marriage, common-law union, adoption or a foster relationship. Anyone who is part of a census family is also part of the same economic family (Statistics Canada, 2022a).
Based on these variables, households are categorized into two groups: households with non-relatives and households with relatives only. Households are classified as having non-relatives if at least two people are not in the same economic family. However, in these households, it is still possible for people to be related, such as two married couples cohabiting. All other households consist solely of relatives.
Financial contribution towards housing costs from household members
To differentiate households based on individual contributions towards housing costs, the household maintainer status is used. A person aged 15 years or older who contributes to regular monthly housing costs is called a household maintainer.Note The information collected in the census reflects only whether a person contributes financially; it does not capture the amount, the presence of a lease agreement or non-monetary contributions.
The number of household maintainers and which household members are maintainers can differentiate households for many reasons. First, economically, more household maintainers indicate a higher degree of combining financial resources. Second, being a household maintainer can be related to improved security of tenure, legally and informally. That is, in some jurisdictions, when housing payments are in relation to a lease agreement, this often affords legal rights against eviction under tenant protection legislation (Canada Mortgage and Housing Corporation, n.d.).
American research on doubled-up households has shown, both quantitatively and qualitatively, different benefits and risks for the host versus the guest family. In the Canadian context, being a household maintainer does not specifically indicate whether the person is part of the host or guest family, as the guest family could also be paying housing costs. Nevertheless, the American literature strongly supports the importance of the financial dynamic between household members.
Based on the household maintainer variable, households are split into two groups: those where housing costs are paid by at least two people not in the same census family and those where costs are paid entirely by one person or one census family. This approach reflects the strong social and economic ties within a census family.
Crowding and core housing need
To differentiate households based on common housing indicators, the crowding and core housing need variables in the census are used.
A household is living in a crowded dwelling if the dwelling does not have enough bedrooms for the size and composition of the household as calculated using the National Occupancy Standard (Statistics Canada, 2022d). Crowding is an indicator often used to identify housing need. It is one of the three housing norms used in Canada to define acceptable housing and core housing need. Australia includes estimates of severe overcrowding in its homelessness statistics (Australian Bureau of Statistics, 2012). The literature on doubled-up households finds overcrowding to be a common and negative experience for families (Bush & Shinn, 2017).
Core housing need refers to whether a private household’s housing falls below at least one of the indicator thresholds for adequacy, affordability or crowding, and the household would have to spend 30% or more of its total before tax income to afford the median rent of alternative local housing that meets all three housing indicator thresholds. The Canada Mortgage and Housing Corporation and housing agencies use the core housing need indicator to develop housing programs that address housing need (Canada Mortgage and Housing Corporation, 2019).
Crowding and core housing need are variables used in practice to identify households with unmet housing need.
Economic hardship
To differentiate households based on economic hardship, different options are available in the census, including housing affordability, poverty status and income.
The housing affordability indicator used in the census is the shelter-cost-to-income ratio. A household is in unaffordable housing if the occupants of the dwelling paid 30% or more of household total income towards shelter costs.
The poverty indicator used in the census is based on the Market Basket Measure (MBM), Canada’s official poverty measure. The MBM is based on the cost of a specific basket of goods and services representing a modest, basic standard of living.
The income indicator used in this paper is total income. It reflects the combined income of the census family or the income of the person when not in a census family, representing the income of the subgroups that would likely remain together if they obtained their own housing.
Housing affordability and poverty have the benefit of being established indicators related to economic hardship. However, they are based on the combined income of the household members or the economic family members, preventing the identification of hardship among individual groups that are cohabiting (especially if they are not combining their income).
Examining income directly allows the economic situation to be assessed for subgroups of household members who are cohabiting. However, there is no agreed-upon income threshold to determine when housing is unaffordable. For this analysis, a threshold of less than $30,000 in 2020 was used. However, higher thresholds could be used to capture more people.
All three measures—housing affordability, poverty status and an income of less than $30,000—are examined in this paper to better understand the financial hardship under different living and cost-sharing arrangements.
References
Australian Bureau of Statistics. 2012. “Methodology for Estimating Homelessness from the Census of Population and Housing.” ABS Website.
Bramley, G., Pawson, H., White, M., et al. 2010. “Estimating Housing Need.” Research Report. Communities and Local Government, London.
Bush, H., & Shinn, M. 2017. “Families’ Experiences of Doubling Up After Homelessness.” Cityscape (Washington, D.C.), 19(3), 331–356.
Canada Mortgage and Housing Corporation (CMHC). 2019. “Understanding Core Housing Need.” CMHC website.
Canada Mortgage and Housing Corporation. 2022 (June 23). “Canada’s Housing Supply Shortage: Restoring Affordability by 2030.” The Housing Observer.
Canada Mortgage and Housing Corporation (CMHC). (n.d.). “Complaints and Evictions.” Retrieved April 9, 2024, from CMHC website.
Frederick, T. J., Chwalek, M., Hughes, J., Karabanow, J., Kidd, S. 2014. “How Stable Is Stable? Defining and Measuring Housing Stability.” Journal of Community Psychology. 42(8):964-979.
Harvey, H., Dunifon, R., Pilkauskas, N. 2021. “Under whose roof? Understanding the living arrangements of children in doubled-up households.” Demography. 2021;58(3):821-846.
Housing and Urban Development. (n.d.). “American Housing Survey Reveals Rise in Doubled-Up Households During Recession.” PD&R Edge.
Lorentzen, J. C., Johanson, G., Björk, F., & Stensson, S. 2022. “Overcrowding and Hazardous Dwelling Condition Characteristics: A Systematic Search and Scoping Review of Relevance for Health.” International Journal of Environmental Research and Public Health, 19(23), 15542.
Office for National Statistics (ONS). 2023 (March 29). “ʻHidden’ homelessness in the UK: evidence review.” ONS website.
Richard, M., Dworkin, J., Rule, K., Farooqui, S., Glendening, Z., Carlson, S. 2022. “Quantifying Doubled-Up Homelessness: Presenting a New Measure Using U.S. Census Microdata.” Housing Policy Debate.
Statistics Canada. 2022a (July 13). “Families, Households and Marital Status Reference Guide, Census of Population, 2021.” Statistics Canada Catalogue no. 98-10-0137-01.
Statistics Canada. 2022b (July 13). “Home alone: More persons living solo than ever before, but roomies the fastest growing household type.” The Daily.
Statistics Canada. 2022c (September 21). “To buy or to rent: The housing market continues to be reshaped by several factors as Canadians search for an affordable place to call home.” The Daily.
Statistics Canada. 2022d (November 30). “Housing Characteristics Reference Guide, Census of Population, 2021.” Statistics Canada Catalogue no. 98-500-X.
- Date modified: