When it comes down to it, there are really only two fundamental metrics in determining whether Vancouver is in the midst of a real estate bubble. The first is land value per square foot in absolute terms.
In Vancouver, it's not that SFH are appreciating faster than multi-unit buildings per say. It's that the land (which is what is actually appreciating) is split more ways on a multi-unit lot. Land value makes up a more significant portion of the value of a single family dwelling. Thus, a run up in land values would lead to SFH housing appreciating more dramatically. This has certainly been the case.
Something striking to consider...
Per square foot land values (which is what should be compared when comparing cities) in the west side of Vancouver are almost universally over $300 and quite often over $400.
That is quite similar to the average land values on Manhattan (save for the most expensive areas around Midtown, etc).
Does it strike you that the economic fundamentals between New York City and Vancouver are similar? Hopefully not. New York is on an entire different level. The GDP of the New York city metro area is similar to the GDP of Canada, and it is one of the largest financial hubs in the world. Manhattan also finds itself with similar, possibly even more restrictive land constraints.
If this doesn't raise some alarm bells you are doing the equivalent of plugging your ears and saying LaLaLa.
The second important indicator is the ratio between land value and rents. As Lio has pointed out, rents are a much better indicator of market demand and the domestic economy. No matter what property prices do, landlords can only charge rents that the market will bear. It wouldn't matter if single family homes were worth $100 million each.. Landlords would still ultimately only be able to charge rent based on demand, which is directly determined by income. So while rent is fundamentally tied to local incomes (the domestic economy), land values are not. Land values can be affected by lending rules, foreign buyers, among other things.
Thus, to see whether land value is overvalued (a classic asset bubble), rent to land value ratios are very accurate in their predictive power. That ratio is more accurate than land value to income in determining whether a market is overvalued.
Rent is a much better proxy for the local economy than income because it takes both demand (determined by income) and supply constraints into account.
When comparing rent in Vancouver to land value, the supply constraint created by geography/ALR is already factored into average rents! So even with the supply argument accounted for the ratio still suggests bubble conditions.
When comparing American cities to Vancouver, it is important to note that rents in American cities are often quite a bit higher. So with the Seattle comparison, you can easily see that after the housing bubble briefly formed then popped, land to rent ratios returned roughly to their historical average and have stayed there.
In Vancouver this ratio was already "off the charts" as of 2011 and it has risen significantly since. For those who aren't familiar, anything beyond 3 standard deviations is considered a significant outlier (although the sample only goes back to 1980 so the standard error probably isn't all that accurate).
Sorry for the long post.
TL/DR: By pretty much every important economic indicator, Vancouver is facing bubble conditions in land values. There is no such thing as "this time it's different", or we are "an exception to the rule". That being said, the bubble will continue to grow until foreign investment slows significantly, so it's not necessarily a risk in the short term.
Prev
Linear Mode
