Much Ado About Nothing: What the Census data say, and don't say, about foreign & temporary residents and unoccupied dwellings
David Baxter, Andrew Ramlo, and Ryan Berlin
The Urban Futures Institute
Since the release of 2011 Census data on housing and families, those of us who are data-inclined have had the opportunity to explore the ways in which our communities changed over the previous five years across a range of dimensions, from the proportion of new dwellings that are being added as apartments to the shifts in family composition and living arrangements.
One specific element that has attracted significant attention of late (at least here in Vancouver) has been the prevalence of housing occupancy by foreign and/or temporary residents; this has been closely followed by concerns over the prevalence of both unoccupied dwellings and foreign investment, specifically as they pertain to the apartment stock in the metropolitan Vancouver region.
Unfortunately, discussions around these issues have suffered from the incorrect reference to—and use of—Census data. Largely a result of not considering the precise Census definitions for the data, the discussions have at best misrepresented what the Census attempts to measure, and at worst supported misconceptions about housing occupancy in the region. Either way, if there is going to be a meaningful discussion about housing, and meaningful policy responses to particular issues or trends that we, as residents of the region, feel need attention, it is necessary to know what the data are and what they mean before we draw any conclusions from them.
In order to provide some clarification on both the definitions and the data, we purchased a custom tabulation of the Census data in question, thereby allowing us to more fully explore and explain two elements of housing occupancy: a) foreign and/or temporary residents and b) unoccupied dwellings.
The following is a summary of the major findings from our research, and it is supported by a technical backgrounder, Housing Occupancy in the City of Vancouver and Region: 2011 Census Results Plus, that details the data sources and definitions in question.
The Census Definitions
Among many other things, the Census counts the number of private dwelling units occupied by usual residents; this includes all units whose residents considered their residence in this region to be their primary place of residence on May 10th, 2011. Units occupied by persons who considered (or who were deemed to have considered) their primary residences to be elsewhere (in Canada or abroad) were classified as occupied by foreign and/or temporary residents. In this group are students who live in a private dwelling during the school year (and perhaps while working at a summer job) but at some point during the year live elsewhere, such as at their parents’ place. This group also includes foreign students, both short term (such as language students) and long term (such as graduate students) who lived here on Census day but have a principle residence elsewhere.
Also included in the foreign and/or temporary group are non-students who have a main residence elsewhere in Canada.This applies to people in a private dwelling in the region on Census day who have a main residence elsewhere in Canada, including those who were working here, on a short term or long term basis; visiting here to see or care for friends or family members; visiting as tourists; and those here on short courses and conferences. Another group or people included in the foreign and/or temporary classification are non-students who have a main residence outside of Canada. This applies to people, regardless of citizenship, who were in a private dwelling in the region on Census day but had a main residence outside of Canada, including tourists, workers, family visitors, entertainers and professional athletes (such as those who play on our home teams, but consider their homes to be elsewhere, and those who play for other teams who happened to be here, and in private dwellings, on Census day).
For these definitions, note a) temporary does not imply any particular length of stay, and can range from over night (in the case of tourists in private dwellings) to year long in the case of students - all temporary means is that people have, or are deemed to have in the case of some students, a main residence elsewhere; and, b) while much is made of the foreign within this group, it is much wider classification as it includes students, workers, friends and family, tourists and some of our sports heroes.
The Census definition of unoccupied units includes much more than units that are vacant on Census day. In addition to units that were empty (without people or furniture), unoccupied also includes all other units that were not designated as a main residence by a Census respondent and in which there were no occupants on Census day. These units range from the vacant and available for occupancy (including newly constructed units for rent or sale, and vacant existing units for sale and rent), to units vacant on Census day but with occupants on their way (e.g., people moving into a recently-purchased home), to occupied by usual residents who were temporarily away and did not complete a Census questionnaire and, finally, to being full of furniture as second residences for people whose main residences are elsewhere. It is important to note that the Census count occurs on May 10th, after the exodus of many students at the April end of the fall/winter academic term, something that has a significant impact on the number of unoccupied units counted in the Census.
What the Census Actually Tells Us
Foreign and/or Temporary Residents in Canada’s Metropolitan Regions
Given recent discussions about dwellings occupied by foreign/temporary residents in this region, it is productive to commence with consideration of how we compare to other census metropolitan areas (CMAs) in Canada. Dwellings occupied by foreign and/or temporary residents in the Vancouver CMA represent 0.8 percent of the dwellings in this region, about the same (insignificant) share as they do in other major metropolitan regions in Canada. In considering the 19 metropolitan regions in Canada with populations of at least 200,000, the Vancouver CMA was right in the middle of the pack with the Saskatoon and Regina CMAs (0.9 percent), and the Montreal and Victoria CMAs (0.8 percent).
The regions in Canada with relatively high levels of occupancy by foreign and/or temporary residents were the Sherbrooke CMA (2.0 percent) and the Kitchener/Cambridge/Waterloo CMA (1.5 percent), both of which have relatively large university and college populations. The Ottawa/Gatineau CMA was third-highest among the major CMAs at 1.2 percent, reflecting the influence of large post-secondary and political/diplomatic populations.
A similar pattern is seen when focusing specifically on the apartment markets of Canada’s larger CMAs. The 1.4 percent of the Vancouver region’s apartment stock occupied by foreign and/or temporary residents sits almost right at the 1.5 percent average for all of Canada’s 33 CMAs. Regions with above-average shares, once again, were Kitchener/Cambridge/Waterloo (3.5 percent), Sherbrooke (2.7 percent) and Ottawa/Gatineau (2.1 percent). There is no evidence in the Census data to indicate anything but normal levels of occupancy by foreign and/or temporary residents in this region: the greater the role played by post-secondary education in a region’s economy, the higher the level of occupancy by this group of residents.
Foreign and/or Temporary Residents in the Vancouver CMA
The Vancouver CMA regional average of 0.8 percent of the housing stock being occupied by foreign and/or temporary residents was not uniformly distributed across either structure types or within the region. Compared to the 0.8 percent average, a below average 0.3 percent of the single detached stock and 0.4 percent of the attached ground oriented stock were so occupied, while an above average 1.4 percent of the apartment stock was occupied by temporary residents. This is to be expected as the apartment market provides a significant share of student’s rental accommodation.
While the City of Vancouver and the District Municipality of West Vancouver were both slightly above the 1.4 percent regional average (with 1.8 and 1.9 percent of their apartment units, respectively, occupied by foreign and/or temporary residents), the University of British Columbia / University Endowment Lands (UBC/UEL) area, with 5.8 percent of its apartment stock occupied by foreign and/or temporary residents, was substantially above the regional average. This spatial pattern clearly shows the impact of post-secondary students on apartment occupancy patterns. There are 250,000 students (including 28,000 international students) registered in post-secondary education institutions across metropolitan Vancouver (along with another 40,000 short-stay international students): if they all lived in one place, they would represent the third largest municipality in the region.
Unoccupied Dwellings in CMAs across Canada
An average of 4.8 percent of the dwelling units in Canada’s 33 CMAs were unoccupied at the time of the Census. With a 5.4 percent level of unoccupied units, the Vancouver CMA was above the CMA average, but the difference was slight compared to other CMAs, such as the Victoria (7.5 percent), London and Windsor (6.9 percent), St. Catherines/Niagara and Sherbrooke (both at 6.8 percent) regions.
Looking specifically at the apartment market, unoccupied units in the Vancouver CMA accounted for 6.2 percent of all apartments here, below the 7.0 percent average for all 33 of the CMAs in Canada. In 14 of the 19 largest CMAs the unoccupied apartment share was higher than here, ranging from 6.5 percent in the Winnipeg CMA to 16.5 percent in the Windsor CMA. While both the Montreal and Toronto CMAs recorded lower levels of unoccupied apartment units — 5.9 percent and 5.4 percent, respectively — they were not significantly below that of the Vancouver region. Given these data, the 2011 Census provides no basis for concluding that there is an excess of apartment units that are unoccupied in the Vancouver region.
Unoccupied Dwellings within the Vancouver CMA
The average of 5.4 percent of all private dwellings in the Vancouver CMA being unoccupied at the time of the Census represented underlying levels of 3.2 percent of the single detached stock, 6.2 percent of apartments, and 6.8 percent of attached ground oriented units. Single detached units accounted for 20 percent of the unoccupied units in the region on Census day, perhaps reflective of 2011’s active real estate sales market.
Within the Vancouver region, with an overall average of 6.2 percent, unoccupied apartments accounted for a slightly above average share in the City of Vancouver (6.7 percent) and West Vancouver (6.9 percent), and well above average shares in Pitt Meadows (8.7 percent), Surrey (9.2 percent), and in the UBC/UEL area (10.1 percent). The spatial pattern of unoccupied apartment units throughout the region is driven by a wide range of factors, from the prominence of student populations within each municipality to sales activity.
The Census Undercount
No discussion of Census population counts would be complete without a discussion of the Census undercount. This is the number of usual residents who should have been included in the Census but were, for some reason or another, not counted. According to current estimates, there were roughly 85,000 usual residents that were missed in the region, 3.7 percent above the Census count reported for 2011. To the extent that these people lived in units that were classified as unoccupied, the actual prevalence of unoccupied dwellings could have been significantly lower than the recorded prevalence as per the Census. Thus, when using Census population numbers, not only must there be an awareness of the fact that university students who left a few weeks earlier are not included (nor were other people who were here but did not have a main residence here), but also that they leave out another 85,000 people.
Much Ado About Nothing
There is nothing in the most recent Census data that provides evidence that the Vancouver region has any abnormal or excessive level of occupancy by foreign and/or temporary residents when compared to other major metropolitan regions in Canada. These data also do not provide any basis for concluding that there is an excess of units in the region that are unoccupied.
Looking within the Vancouver region, while the City of Vancouver is slightly above the CMA-wide average for both units occupied by foreign and/or temporary residents and unoccupied units (apartment and otherwise), there are other parts of the region where such occupancy is much more prevalent, most notably the UBC/UEL area. Within this context, it is important to recognize not only the underlying characteristics the dwelling stock, but also of the residents who may occupy it (e.g., the Census treatment of students’ places of residence).
Further to this, the Census provides no indication of the degree to which foreign and/or temporary residents were not Canadian residents or their actual citizenship; no indication of the duration of temporary residency, which could be for as long as eleven months in the case of students who only stay with their parents between school and job; no indication of whether the temporary residents were owners or tenants; and no indication to the degree to which unoccupied units were vacant rather than temporarily unoccupied on Census day. There are no Census data that apply to discussions of foreign ownership or investment in housing, and none that apply to foreign occupancy, except to the extent that persons with a main residence outside of Canada are included; but note that along with this group are the post-secondary students whose parents homes are outside the region of study, and other people whose main residence is not their residence in this region.
The bottom-line is that the 2011 Census data clearly show that much ado is being made about nothing when it comes to the prevalence of foreign and/or temporary residents and unoccupied dwellings in this region. There are significant housing issues in this region – the levels of occupancy by foreign and/or temporary residents and level of unoccupied units are not among them.