Профиль

Courageous. Principled. Leadership. In this economy?

Everyone loves large planning exercises like the Vancouver Plan. But when it's time for implementation people get cold feet. Even though the Villages is probably the mildest portion of the overall plan. Will the Vancouver Plan suffer the same fate as City Plan in the 1990s and just fizzle?

Great tribute to Michael Coyle (@oplopanax.ca) up at CBC. Surely just a start. He'll be sorely missed. www.cbc.ca/news/canada/...

I was intrigued enough by this measure to do some further analysis, including looking at national data going back to 1940 and at a bunch of metro areas since 2005. I wrote about it here: domainexpansion.substack.com/p/an-easy-an...

Another way to get at this is with DiPasquale-Wheaton's age-expected households. It can't be used directly to estimate shortages, but it's easy to construct and shows changes across time and across geographies and gives a very similar picture.

I think that's a useful way of doing this. One can get a little more precise using "Minimal Household Units". And it gets more interesting when doing this at the metro level. And correlating it with rents.

He fit the same statistical models with three different software and got much different estimates. It’s another dimension of the multiverse. statmodeling.stat.columbia.edu/2026/07/27/5...

I am so sorry to hear that, my heart goes out to you all.

* Once you disable the text message reminders with their predatory charges.

There is the question of spillover, so spatial or other effects across the DID boundary, but then there is the within region spatial effects. They don’t share code but it looks like they treated each DA as an independent unit. Which is wrong. That artificially reduces standard errors.

Two other red flags: * Claim Dissemination Block "coding systems change between years" -- true but applies the same to Dissemination Areas, makes me wonder if they missed accounting for that * They deal with fine geography spatial data but no spatial methods used. Standard errors are uniterpretable.

Formally, these are "homes occupied by temporarily present persons". And that matters n the COVID context where we saw a large dip in exactly these that was entirely driven by the drop in homes occupied by temporarily present persons as universities were remote and international migration dipped.

More importantly, the "usual place of residence" concept codes a good share of students and a chunk of non-permanent residents to live elsewhere from where the census found them. E.g. living with parents instead of near their place of study. The homes they live in show up as "empty" by their count.

They also include moving vacancies that aren't "on the market", it's not uncommon for people to pay rent in two places at the same time during moves. And those scale with mobility. Homes under extensive renovation also factor. Both of these arguably changed during COVID, their timeline endpoint.

I would add that the vacancy rate comparison is very problematic. They don't actually look at "empty" or "unoccupied" homes, they look at homes "not occupied by usual residents" on census day. There are several issues with that beyond just second homes.

Mini-thread on the Vancouver vacancy tax (and resulting study). The tax is Vancouver's 2017 empty homes tax; the study from CD Howe looks at its impact. The study claims the tax had large impacts on vacancies and minimal impacts on prices. As I'll discuss, I think there are problems with the study.

New blog post: No, Manchester did not build more homes than Greater London last year jamesjgleeson.wordpress.com/2026/07/26/n...

The official announcement by the misery does not address the question of cycling and pedestrian access at all. Not unexpected, but still disappointing. news.gov.bc.ca/releases/202...

It’s my commute to work. They better accommodate cycling along there, the 8th ave car tunnel is a really poor alternative. As is the 4th ave race track.

GHSL data is great, endless fun.