The Hongcouver

In Bizarro Vancouver, 25,000 households declare less income than they spend on housing alone

The unusual phenomenon exists in one in ten households across city – with the situation extreme in Arbutus, South Granville and other wealthy neighbourhoods

PUBLISHED : Thursday, 20 August, 2015, 1:39am
UPDATED : Tuesday, 23 February, 2016, 11:49am

Vancouver watchers are getting increasingly used to Bizarro-type statistics about their city. How could they not, when the average detached house price is about C$1.4 milion?

But I came across a true head-spinner this week, courtesy of mathematician Dr Jens von Bergmann, a former teacher at Notre Dame and the University of Calgary who now runs MountainMath Software, a small data firm based in Vancouver.

According to Statistics Canada data from the 2011 National Household Survey that was mapped by von Bergmann, there are 24,960 households in the City of Vancouver where the amount spent on shelter (mortgage, rent, utilities, property tax and strata/condo fees) exceeds 100 per cent of their declared household income. Think about that: these households claim to earn less than they spend on accommodation alone.

And it’s not an insignificant cohort. They represent 9.5 per cent of all households in the city.

But you might want to hold off on the pity party - while some may be undergoing financial stress, most of the neighbourhoods where the situation is extreme aren’t exactly struggling.

Fascinating interactive maps devised by von Bergmann make this clear . While there are neighbourhoods with relatively high proportions of such households (say, 20 per cent or greater) scattered across the city, they are conspicuously concentrated in the city’s expensive Westside.

In some neighbourhoods – technically, census dissemination areas – the proportion of households in which shelter costs exceed entire income rises above 40 per cent. These neighbourhoods blare red like sirens on von Bergmann’s map.

To be clear: this data compares an individual household’s actual shelter costs with the same household’s declared income. It’s not based on averages or medians across each dissemination area.

For instance, in dissemination area 59150581, a cosy pocket of Arbutus Ridge nestled between Arbutus and Macdonald streets, 44.8 per cent of households (65 out of 145) claim to earn less than their shelter costs. The median dwelling value in this apparently impoverished enclave? C$1.98 million. Median individual income from all sources? C$19,993.

Another standout neighbourhood is dissemination area 59150507, in exclusive South Granville, where 43 per cent (65 out of 150) had shelter costs exceeding total income. Median dwelling value, C$1.80 million. Median income, C$13,572.

And in that South Granville area, home ownership was 87 per cent; in the Arbutus dissemination area, it was 100 per cent.

Things are a bit different in Yaletown’s dissemination area 59153851, between Hornby and Howe, where 50 per cent of households spent more on shelter than they supposedly earned. As you’d expect in condo-heavy downtown, median dwelling value is much lower (C$349,900) and the proportion of renters much higher at 73 per cent. Median income is C$15,098.

There’s a whole new world of weird in dissemination area 59153466 - which abuts Coal Harbour and includes the swanky Fairmont Pacific Rim condos and hotel - a stunning 62 per cent of households claim to have lower income than shelter costs. Not only that, but a massive disparity between average and median individual incomes suggest that a large proportion declare virtually no income at all: the average is C$14,293, but the median just C$943 per year.


Astronauts and other absentee earners

To put all this in context, the relative presence of households spending more on housing than total income is far lower across Canada at about 4.4 per cent (meaning that Vancouver’s rate is, proportionally, 116 per cent greater). In Greater Toronto , it’s 5.9 per cent; Montreal  5 per cent and Victoria 5.4 per cent.

So what are the possible explanations for Vancouver’s excess?

The data is certainly consistent with immigration patterns to the city, which has received a globally exceptional number of millionaire migrants (50,000-plus individuals in the past decade, representing 14,000 plus wealthy households) who typically declare extremely low income.  It doesn’t prove categorically that these rich immigrants – most of them Chinese of late - are using foreign-sourced wealth or non-declared foreign income to fund housing choices in Vancouver (which might also imply tax cheating). But I can’t think of a viable large-scale alternative explanation. It fits the ethnic Chinese “astronaut family” narrative.

Playing devil’s advocate, could it be instead that Vancouver’s high shelter cost:income ratios are explained by a high presence of aged retirees, whose housing costs are borne by locally-earned savings, and not income?  That might apply to some, but overall it holds no water, since residents of the City of Vancouver are substantially more likely to be of working age than Canadians are as a whole, 74.6 per cent versus 68.5 per cent nationally.

Could it be that Vancouver has an abnormally high number of households in which shelter costs are funded via ongoing earnings from elsewhere in Canada – say, an absentee oil-patch worker? This thesis requires some logical leaps: assuming that such people exist here in large numbers (and I’m not aware of evidence that they do), why would they behave in such a consistent fashion that they would form distinct geographical enclaves across the city?

On the other hand, it’s demonstrable that ethnic enclaves do exist. In our Arbutus and South Granville dissemination areas, for example, the proportions of residents speaking a Chinese mother tongue are abnormally high at 51 per cent and 47 per cent (compared to 15 per cent across metro Vancouver).

The foreign-income theory doesn’t apply in every single case, such as 59153466 near Coal Harbour, where the proportion of Chinese speakers is rather low at 13 per cent, and which von Bergmann suggests may be populated by service-industry staff relying on tips and unreported earnings (hence the ultra-low median income). An extremely high proportion of households there (44 per cent) are deemed “unsuitable” by Statistics Canada, reflecting a high number of residents per room, which could fit von Bergmann's theory, as could the huge diversity of non-English languages spoken in the neighbourhood.  

Von Bergmann also points to enclaves near the University of British Columbia. He suggests these are student households whose costs are covered by parents living elsewhere - and not necessarily overseas. Such circumstances would account for many across Canada.

“But also, if you’ve come in on a [wealth-determined] immigration programme, you could be just living on your savings. That’s definitely something we should be seeing in the data [too],” he said.

(Note: This article previously referred to dissemination area 59150581 as being in Dunbar, instead of Arbutus. For anyone checking Bergmann’s figures, note that Statistics Canada rounds data for anonymity, throwing out some totals by a small amount. A map depicting the simple number of households with income lower that shelter costs which helps illustrate the city’s east-west divide, can be seen here)


The Hongcouver blog is devoted to the hybrid culture of its namesake cities: Hong Kong and Vancouver. All story ideas and comments are welcome. Connect with me by email [email protected] or on Twitter, @ianjamesyoung70. RSS feed here.