Thursday, September 12, 2013

Vancouver CMA Population and Dwelling Distributions 2006 to 2011

Statscan released their 2011 National Housing Survey results earlier this week. I found a couple of datasets that are broadly comparable to the previous 2006 census data. Below are graphs looking at overall population growth and dwelling composition based on structural type (defined here). Here is a brief summary of some structural type definitions:
  • single detached: one dwelling with no walls adjoining another structure
  • apartment, duplex: detached housing with suites (for example)
  • row house: dwellings in a group with shared walls
  • apartment >= 5 stories: condo towers
  • apratment < 5 stories: low to mid-rise condos
The first graph looks at the average household size by structural type with the change since 2006. Notice that "apartment, duplex" and apartments with >= 5 stories saw increasing household size. Row and single detached were unchanged, and low-rise apartments saw lower household size. Overall the household size was unchanged since 2006. (Click on image to enlarge.)

Next is looking at the distribution of dwelling growth based on structural type. On a per-dwelling basis apartments were the most significant contributor to dwelling formation however on a population growth basis the at-grade structures (single detached, apartment duplex, and row house) housed the majority of the region's population growth.

As to why this is, we can look at the distribution based on household size to see that while apartments contribute more on a per-dwelling basis their household sizes are smaller so will have a lower impact on population absorption on a per-unit basis. This is obvious but for posterity here are the distributions:

The most interesting thing for me is the stagnation of the household size number since 2006. A few other notes:
  • 67% of the metro's population growth was absorbed in single detached, apartment duplex, and row housing.
  • Households of 3 or more people made up 39% of dwelling formation but 62% of population growth.
Household size could be bottoming earlier than predicted. There are, however, some plausible reasons why household size has only temporarily plateaued instead of continuing a downwards trend since the 1970s:
  • The 2006-2011 period encompassed a severe recession and slow recovery with elevated unemployment. This could cause an increase in household size, for example an unemployed worker moves in with parents, or a student decides to live at home instead of moving out of the family home.
  • The NHS has collected on a voluntary basis so there may be errors that skew the 2011 data.
  • Rents increased above inflation over the 2006-2011 period and that could cause some increase in household size.
On another interesting note, laneway houses are likely counted as "single detached" dwellings. There were about 200 such dwellings build in the City of Vancouver before the 2011 NHS. Assuming laneway starts average 40 per month for the next three years, this could add about 1800 additional "single detached" dwellings to the dwelling count for the 2016 NHS.

No comments: