There is no Brain Drain, but there might be Zombies

co-authored by Jens von Bergmann & cross-posted at MountainMath & (as of Feb 8th) updated with slightly better mortality estimation

 

Zombie attack! Zombies fleeing Vancouver want to eat your brain… drain… or something.

A couple of weeks ago The Canadian Press reported a story asserting that young professionals were leaving Vancouver because of the high cost of housing. This fits in with a common zombie refrain that we hear from the media. It’s a story that just won’t die, no matter how many times it’s proven wrong: Millennials, or young people, or boomers, or people important for some other reason are leaving Vancouver because of housing. Usually there are supporting anecdotes, and indeed, it’s not too hard to find people leaving Vancouver who will tell you about their frustrations with housing. But here’s the thing: there is almost never supporting data that actually indicates a decline in people worth caring about. Why? Two reasons. First, in growing cities, like Vancouver, when some people leave, even more people come in to replace them. Second, ALL people are worth caring about.

If we set aside that ALL people are worth caring about – just for a moment – we can take up some important questions about differences in in-flows and out-flows of people in Vancouver. Maybe there are aspects of in-flows and out-flows that should trouble us. In The Canadian Press story, we’re led to believe Vancouver is experiencing a brain drain, so that all the smartest and best people are somehow leaving and they’re either being replaced with people who are not so smart OR they’re not being replaced at all. As noted above, Vancouver is growing. So we know whoever leaves is being replaced, and then some, by new people coming in. But are the people arriving in Vancouver somehow less brainy than those leaving? We’re both immigrants to Vancouver, and quite frankly we find that a little offensive. Everyone arriving in Vancouver has a brain, so population growth cannot result in a brain drain. But we set aside, for a moment that idea that ALL people were worth caring about. So let’s try putting differences in in-flows and out-flows in slightly less offensive terms by returning to the “young professional” framework. Are people arriving in Vancouver unable to do the same kind of professional work as those who leave? Are we losing out on educational credentials?

Ideally we could easily access direct information on in-flows and out-flows to Vancouver (and in some places with population registry data, this is easily accomplished). In Canada we work mostly with census data, and the out-flow data, in particular, isn’t generally made public. But as we’ve demonstrated previously, we can compare across censuses to get net migration data broken down by age group. We just age people forward from one census to the next and compare how many we see in the next census to get a sense of how many people – in net terms – must’ve moved in or out over the years in between.

Now if we’re interested in education then it complicates age-based net migration models. After all, people can and do acquire new educational credentials as they age forward in time. That said, we can probably assume that most people who acquire university degrees and more advanced credentials do so by age 25. We’ll leave out some late achievers, for sure, but if we assume we have a pretty stable division into those with a completed Bachelor’s degree or more, and those without by age 25, then we can get a sense of how those populations change as they age forward in time. So, with apologies to late achievers, that’s what we’re going to do.

We’ve got ten year age groupings by education to work with in 2016 data. So let’s go back to 2006 data for comparison. Is it plausible that we lost a bunch of “young professionals,” defined as people with university degrees, who weren’t replaced as they aged forward and left Metro Vancouver between 2006 and 2016? Data says… nope.

Prof-Updated-Mortality2

As a matter of fact, Vancouver added a lot more young university graduates than left. Young people with university degrees continued to arrive in greater numbers than they left well through their thirties and on into their forties (we like to think of forties as young). The age labels here refer to people’s “in between” age, that is the ages they mostly passed through between 2006 and 2016 (i.e., the age range each group was in 2011). It’s only once those with university degrees hit their fifties that we start to see a roughly even net flow out of in Vancouver. What’s more, this pattern looks very similar in other major Canadian metro areas. The only exception is Montreal, where people with university degrees really do stop arriving in their forties. But it’s probably not a housing crisis driving them out.

Strikingly, across the board, young people with university degrees are far more likely, on net, to move into our major metro areas than people without university degrees. In many respects, we should expect this. Professionals, in particular, are often drawn by their economic opportunities. Once they arrive anywhere, they’re often paid well enough that they have an easier time navigating local housing markets than non-professionals. Yes, professionals may also have higher expectations about what kinds of housing they deem acceptable than others, but people adapt. One of us has written a book with that theme. In the same way that professionals may drive gentrification, professionals are actually at LESS risk of displacement out of expensive places, like Vancouver, than are non-professionals.

Let’s double-check the results for Vancouver by looking at in-flow data. The Census provides information about where people lived five years before arriving at their current destination. Do we really see a lot of professionals moving into Vancouver through their thirties and forties? Yes. In fact, for “Skill Level A Professionals” this is exactly what we see. We don’t know how many are leaving from this data, but we know a lot of professionals are arriving – more so than in other occupational skill-level categories.

occupation-inmove-2

For mobility data the age group labels refer to people’s age in 2016. For an alternative view we can group non-movers and non-migrants (people that did move but not to a different city) together and show the makeup of each skill level by mobility and age group. Again we see that professionals tend to have higher shares of migrants than other skill levels, especially in our lower two age brackets. Those in occupations requiring only a high school degree or on-the-job-training are actually the least likely to come from afar.

occupation-inmove-3

Takeaway: we do not have to worry about a “brain drain” in growing cities like Vancouver. Moreover, we don’t have to worry about professionals leaving. Due to better pay, professionals are better equipped to deal with a tight housing market than most others. Building more housing would certainly give professionals more options to choose from, and we might want to relax our millionaire zoning to direct professionals toward competing with the independently wealthy rather than the poor and working class. But it’s the poor and working class we should really be worried about losing. More housing can lead to a more equitable city with room for people who aren’t well-paid professionals or independently wealthy. And if we want to prevent displacement, we should focus more on those actually at risk. That suggests both building more and promoting a LOT more non-market and rental housing.

Methods

There are some details to be explained when computing net migration data for professionals. We already noted that professionals might get degrees at some later stage in life, but that tends to bias our estimates toward lower professional in-migration. Furthermore, when computing net migration one needs to kill off an appropriate number of professionals to account for mortality as Nathan has explained in details before. We use BC mortality rates for the appropriate years and age groups for this, but that probably over-estimates mortality as educated people tend to have lower mortality rates. This would bias our estimates toward higher professional in-migration. We could adjust for that by reading into the literature to figure out the appropriate fudge factor, but the effect is so small that we just ignored this. We made some adjustement to how we compute mortality rates and now assume a 20% reduced mortality rate for people with bachelor or above, and according higher mortality rates for people below a bachelor. This is a very rough approximation of the impact of educational attainment on mortality.

Those interested in even more details we direct to the code for the analysis, where Jens is teaching Nathan how to code with R.

How are condos used?

Comparing How Condos are Used Across Canada

Co-authored by Jens von Bergmann; Nathanael Lauster; Douglas Harris (Cross-posted at mountainmath.ca)

Condominium apartments are fascinating! At their heart lies a relatively recent legal innovation enabling individual ownership of units in multi-unit developments. Since their arrival, condominium apartments have become places to build homes, sources of rental income, sites of speculative real estate investment, and experiments in private democratic government. They’re also in the middle of many on-going debates about housing and the future of cities in Canada and around the world. In 2018, we formed a team to study condominium apartments and how they were being used in order to better inform public and academic debates. Team members include data analyst and mathematician Jens von Bergmann, sociologist Nathanael Lauster, and law professor Douglas Harris. We recently presented some preliminary findings at the National Housing Conference in Ottawa and we’re looking forward to continued research collaboration.

Here we make public some basic information about the development and use of condominium apartments across different metropolitan areas in Canada.

unnamed-chunk-2-1

The first thing to note is that the legal architecture of condominium is deployed across a broad range of structure types. In addition to apartments, developers commonly use the condominium form to subdivide row houses, and occasionally single-detached houses (as in some gated communities). Nevertheless, condominium is used most commonly to subdivide ownership in low-rise and high-rise apartment buildings, and that’s what we focus on here.

The next thing worth noticing is that condominium is much more common in some metro areas than others. Vancouver jumps out for the proportion of its apartments – and housing stock overall – owned within condominium. Calgary and Edmonton also rely heavily on condominium to subdivide apartment buildings, although these sprawling metro areas are dominated by single-detached houses, much more so than Vancouver, reducing the overall prevalence of condominium.

We know that condominium apartments are exceptionally flexible forms of housing, but how are they being used across different metro areas? What proportions are owner-occupied? Rented? Occupied temporarily? Unoccupied?

We couldn’t extract data to answer the last two questions from the census because condominium status is recorded by respondents. However, using a variety of datasets, we figured out a transparent and replicable (if somewhat complicated) method for estimating temporarily occupied and unoccupied condominium units.

The answers to these questions about how condominium apartments are used speak to important elements in popular discourse and public debate. Since provincial governments introduced a statutory form of condominium in the late 1960s, developers have built condominium buildings rather than purpose-built rental apartments across much of Canada. Does this also mean that the proportion of owner-occupiers increases while that of renters decreases in cities where condominium developments proliferate? Or do owner-investors rent out their condominium units, augmenting the existing rental stock?

condo_usage-1.png

Our findings on how condominium apartments are used are really interesting! In all the metro areas we analyzed, the modal use of condominium apartments is owner-occupation. As a result, it appears that condominium apartments are enabling more homeowners to live in increasingly dense cities.

However, condominium apartments also make up a substantial proportion of the rental stock in many metro areas. While many condominium apartments are rented, relatively few show up as vacant (i.e. empty but listed as “for rent”) at any given point in time. Here we distinguish these rare vacancies, which are good for renters, from unoccupied condominiums. In tight markets such as Vancouver and Toronto we see effectively non-existent condominium apartment vacancy rates, comparable to purpose-built rental vacancy rates.

The least common use of condominium apartments is as a temporary residence (where owners declare their principal residence as somewhere else in the census, but occupy the unit occasionally).

Finally we get to the “empty condos,” or those that show up as unoccupied in the census. Overall, we estimate that between 10% to 23% of condominium apartments were unoccupied in 2016, depending upon the metropolitan area. We don’t know why so many condominium apartments appear to be unoccupied, but it likely relates to their newness and to their inherent flexibility as property. Flexibility can show up in the census as “unoccupied” directly, as when owners use condominiums as second homes, and indirectly, as when condominium apartments are left empty in order to facilitate transactions between uses. We suspect that condominium apartments may cycle more frequently than other forms of property between different uses and occupants, thus creating transition periods without occupants and inflating the proportion of unoccupied units. For instance, condominium apartments can more plausibly be re-claimed for owner’s use than purpose-built rental apartments, cycling in an out of rental supply and potentially creating less stable rental housing.

Strikingly, Vancouver and Toronto stand out as having the lowest proportion of unoccupied condominium apartments, a finding that may be somewhat counter-intuitive given the public attention that vacant units have received, rightly or wrongly, in both cities. When metropolitan areas rely upon condominium apartments as a key form of new housing supply, they should take the flexibility of the form into account. However, it appears that the proportion of unoccupied units in the housing stock will rise as the proportion of condominium apartments in the housing stock increases because condominium apartments are more likely to be unoccupied than purpose-built rentals, a pattern also noted with respect to other flexible housing forms, such as secondary suites (especially basement suites, which show up as units in a “duplex” in the census). This means that even though a smaller proportion of condominium apartments are unoccupied in Vancouver than elsewhere in Canada, a larger proportion of Vancouver’s housing stock shows up in the census as unoccupied.

In Canada’s three largest metropolitan areas, a pretty simple rubric applies: for every ten condominium apartments built, six are owner-occupied, three are occupied by renters, and one is unoccupied. In Calgary and Edmonton, add a renter and take away an owner-occupier. The data for the other cities we surveyed is available in the graphic above. As a bonus, we also provide a comparison with estimations from 2011 data to show changes over time in the graphic below.

2-1

In Vancouver, where condominium apartments have been an established part of the housing market for longer than in the rest of the country, there is very little change in the occupancy pattern between 2011 and 2016. In other big metropolitan areas, it appears that condominium apartments are increasingly used as rental stock. In most cases, the proportion of empty condominium apartments appears to be decreasing, something that may reflect the lingering effects of the 2008-09 property market crash. However, this is all very preliminary. But we’ll keep looking at the details as we proceed!

Methods

We mixed two data sources to arrive at these estimates–the Census and the CMHC Rental Market Survey–and that made coming up with the estimates a little more complicated. There are several assumptions that go into the estimates, and there are several issues with mixing the data that we set out below.

Overview

We cut the condominium stock into five different categories. The numbers of units occupied by owners and renters are straight-up census estimates from 98-400-X2016219 and 99-014-X2011026. To estimate the unoccupied units and the units occupied by temporary residents we used a custom tabulation of Structural type by Document type. We received this cross tabulation from Urban Futures, which one of use has worked with before on secondary suites. Both of those variables–the categorization of the dwelling type as well as the decision to label a unit without a census response as empty or occupied by someone who did not respond–is made by the enumerator. This allows us to ascertain the structural type of unoccupied units, and we can also get that information for units that are temporarily occupied.

So, we know how many apartment units were classified as unoccupied or temporarily occupied. To estimate how many condominium units fall into that category we need to make some assumptions. First, we assume that the apartment stock consists of three distinct type of units: condominium units, purpose-built rental units and non-market housing units. That’s not quite accurate. For example a single-family home with two secondary suites will be classified as an Apartment, fewer than five storeys if the census found the suites. These do exist in Vancouver, and elsewhere, but their numbers are small.

Given those three types of apartment units, we need to understand how many of the unoccupied and temporarily occupied units fall into each category. The CMCH Rental Market Survey has annual estimates of vacancy rates and universe size for the purpose-built rental stock. We take those estimates, only counting apartment units, to attribute unoccupied units to the purpose-built rental stock. In Vancouver, with its extremely low vacancy rates, this is a fairly small number. In Halifax, that number is comparatively larger. Further, we assume that the non-market units have a vacancy rate of zero, so that there are no empty non-market units. What’s left over we assign as empty condominium apartments.

Finally, we use the estimate of vacant condominium apartments and those on the rental market from the CMHC Secondary Market Rental Survey, using their estimates of the condominium vacancy rate and the condominium rental universe. The vacancy rate is not available for all years and all CMAs. We have marked the CMAs with an asterisk in case the data was not available and back-filled it with our estimate of the condo rental universe and the Rms vacancy rate. We have seen previously that the Rms vacancy rate tracks the secondary market vacancy rate reasonably well.

Attributing the temporarily occupied units gets even harder, but the numbers are smaller so getting things a little wrong has less impact. Here we again assume that no temporary residents live in non-market housing, and we assume they are equally likely to live in a condominium apartment (as owner or renter) or rent in purpose-built. That is a bit of a judgement call, but the details of these assumptions don’t make much of a difference to the numbers, and we invite people to grab the code if they would like to adjust the assumptions.

There are several issues when mixing CMHC Rms data with census data. For one, both are point-in-time estimates for slightly different times. The census is pegged in early May, the Rms for October. There may be fluctuations in temporary and unoccupied units, in particular in areas dominated by universities such as Waterloo, with the census being outside of the regular semester and the CMHC survey within.

Next comes the geographic problem, with CMHC switching to new census geographies at the end of the year, so the rental universe still reflects the previous census geography. Montreal is one such example where the CMA changed 2011 to 2016 as we have explained before. That leads to problems when estimating the rental universe, but the effect is moderated when focusing on the empty units.

Another issue is that the definition of apartment that CMHC uses differs slightly from the census.

Finally, for estimating the vacant condominium apartments that were on the rental market we used the CMHC rental condo universe estimate and not the one we derived from the census. There appear to be some differences in how CMHC and the census estimate rented condo units, with CMHC relying on surveys of property managers. In BC that likely involves tallying up units for which Form K was filed, likely leading to CMHC under-estimating strata rentals.

It is instructional to compare the two different estimates.

unnamed-chunk-7-1

With the exception of Hamilton, the census condominium rental estimates are higher, in some cases substantially so. To shed more light on this we also compared the estimates of overall condominium apartments.

unnamed-chunk-8-1

We looked at two separate census estimates: the occupied (by permanent residents) units that come straight from the census by filtering occupied units for apartments that are stratified, and the overall condo estimate that we derived by adding in vacant and temporary units. With the exception of Montréal the census estimate of occupied units only comes quite close to the CMHC condominium universe estimate. The differences are worth looking into in more detail at some point.

Waffle graphs

To communicate the makeup of condominium apartments we settled on a custom version of a waffle graph. Displaying proportions on a square grid makes it easier to read them compared to pie charts or tree graphs. The 10×10 layout rounds numbers to percentage points, which is the appropriate level of accuracy given the uncertainty in the data and is intuitive to understand. When rounding to the nearest percentage, the numbers don’t always add up to 100. So we don’t do traditional rounding but round with the constraint that the total adds up to 100 while minimizing the \(l_\infty\) error.

This does introduce potential problems when comparing across time or across geographies, where theoretically we could see an increase in the number of squares in one category although the actual estimated share dropped. This will only happen under very specific circumstances, and we checked that this did not occur in our graphs.

Reproducibility

The code underlying this post is available on GitHub, as are the parts of the custom tabulation for 2016 and 2011 used in this post. Part of the Statistics Canada data we used requires conversion from XML into more manageable data format which, for performance reasons, requires python to be installed next to R that runs the rest of the code.

A modest proposal regarding community consultation in Vancouver

All too often our elected officials ignore vital input from our communities simply because, as they note, “we’re not looking at doing a full-blown consultation process.”

We believe full-blown consultation is always necessary for everything. But how do we do it in a way that enables places to speak? How do we insure we’re only hearing from the right voices, people who are actually from the area? After all, shouts and demands are too often heard from the wrong side of the wall outside our neighbourhoods, diminishing the rightful voices of those within, We’ve kept those people out for a reason! diminishing all that we love about our neighbourhoods.

Castle Community consultation should only involve those people who actually live within the walls community, and we mean own property really live there. Furthermore, only long-time residents should be consulted, by which we mean dynastic residents who can speak for their enduring ties of place. How long is required to get place in your blood? Seven or eight generations should suffice. That’s why we advocate requiring a blood test for all of those who participate in community consultation. Only those with a minimum of 10% place in their blood should get a voice.

By this we mean to signal that both newcomers and prospective newcomers should know their places and not endeavor to speak for our places. They cannot speak for community. Most of them are only here at all because homes have been turned into commodities, for sale to the highest bidder!

1906-Lots-for-Auction-Vancouver-Archives
Shameful commodification of Vancouver homes

Can you imagine? One cannot auction a home! Think of the speculators and profiteers! The only homes that are really homes are those that have been passed down between generations, or perhaps provided to the servants established as non-market somewhere else without raising taxes with appropriate community consultation. If it’s not yours by birth or royal fiat, then it’s not really yours at all.

For similar reasons we believe we’re also justified in limiting the language of consultation. If you and your lineage can’t be bothered to learn the local language here then you demonstrate insufficient ties to consult on the future of place.

All consultation in Vancouver should henceforth take place only in English one of the two official languages of Squamish or Halkomelem.

Sincerely,

A concerned citizen

 

[ed. note: the above is satire directed at a certain strain of community consultation and settler nativism. But providing much more meaningful consultation regarding development with local First Nations would be an extremely good thing, and I’d love to learn Squamish or Halkomelem!]

Checking in with Numbeo

For those interested in making international comparisons concerning rents and housing prices, Numbeo is a potential god-send. I say potential, because there are still some big data quality concerns. But the basic idea is sound: crowd-source estimates of rents and housing prices (as well as costs for all sorts of other things), both for the “centre” of cities and farther out. The end result is a real competitor to even iffier rankings for things like quality of life (looking at you Economist Intelligence Unit!) I’ve been playing around with crowd-sourced data again recently, so I was reminded of Numbeo and thought I’d take a look.

How is Numbeo holding up? And what can it tell us about current housing dynamics? First let’s see what Numbeo tells us about Vancouver, based upon 18 months of crowd-sourced data from 93 contributors (as of Oct 24, 2018):

RENT: Numbeo estimates that rent for a 1BR in the centre of Vancouver average about $1930.86 (CAD). This compares nicely to a listing informed estimate of $1950 for Vancouver 1BRs from Louie Dinh (confirmed as approximate for Downtown unfurnished apartments by a scraper who shall remain anonymous). This runs high compared to CMHC Rental Market estimates of rents for Downtown Vancouver ($1468), but that’s to be expected given that CMHC includes all renters, including long-timers protected by rent control. That said, the CMHC’s estimate for Downtown Condos rented out ($1900) is a lot closer (see p. 35 of report).  All things considered, Numbeo estimates strike me as reasonable for current rents on offer given the vague parameters (Vancouver centre).

PRICE: Numbeo estimates price per square foot for an apartment in the centre of Vancouver at $1,091/sqft. Looking around, this compares pretty reasonably – if a little low – with recent RE/MAX estimates ($1,195/sqft) and even better with realtor Steve Saretsky‘s handy reporting for Sept 2018 ($1,026/sqft). Worth noting that some lag may be expected given the 18 month reporting period from Numbeo.

INCOME: Numbeo estimates an average monthly net salary (after-tax) for Vancouver of $3,170/month. Looking at the Canadian Income Survey (CANSIM 11-10-0238-01), the average monthly after-tax income in 2016 was estimated at $3,042/month, and it’s surely gone up since. Again, seems pretty reasonable as an estimate.

I think it’s worth continuing to check in on Numbeo estimates, which may also vary dramatically from place to place, especially since the number of observers doing the crowd-sourcing also varies a lot (only 18 in Albuquerque!). But on the whole, Numbeo seems to be doing ok for Vancouver, the city I know best.

COMPARISONS: So if Numbeo data seems to be doing ok where I know it best, let’s do some comparisons! Here I provide some basic data for selected North American cities from Numbeo on one-bedroom apartment rents and price per square foot of apartments centrally located in select cities. From here on out, everything is reported in US dollars (just because it made things a little easier).

Price-Comparison-Numbeo-Oct-2018-B1

Cities are ordered by 1 BR centre rents, and the extreme high rent American cities – San Francisco, New York – lead the pack. It takes awhile to get to a Canadian city, starting with Toronto (right after Nashville!) before hitting Vancouver. After that, I pick out a few more of the big Canadian cities. I also add places like Honolulu (expensive resort city) and Albuquerque (one of my home cities!), just for kicks, and low-rent Montreal rounds out the pack at the bottom. Rent and price are correlated (r=0.84), but not perfectly. Strikingly, compared with American cities, all of the Canadian cities have higher prices than one might predict based upon their rents. Of cities examined here, Vancouver ranks 4th highest in price, but 17th in rents.

What happens if we add incomes into the picture? Below I take the same cities and divide both rents and prices by incomes to get simple estimates of relative housing costs. Now the familiar (to Vancouverites) pattern emerges of Vancouver being the priciest real estate in North America, followed by New York and Toronto. Canadian cities look pricey in no small part because our after-tax incomes look relatively low compared to Americans. Rent-wise the story is a bit different. New York and Miami lead the continent, followed by Vancouver, Toronto, LA, San Francisco, and Boston, all hanging reasonably close together.

Price-Comparison-Numbeo-Oct-2018-B2

Relative to income, there’s no doubt both Vancouver and Toronto are expensive places to live in North America. But these are also places with a lot of international immigration. Immigrants make up nearly half the population of Toronto (46%), followed closely by Vancouver (40%). And immigration is increasingly Asian, especially in Vancouver. As I’ve pointed out before, it’s also useful to put Vancouver – in particular – in the context of the broader Pacific Rim.

Here’s base rents and prices (USD), drawn from Numbeo.

Price-Comparison-Numbeo-Oct-2018-B3

San Francisco still leads by rent, but it’s got nothing on Hong Kong when it comes to price. Notably, the price per square foot for apartments in Shanghai, Beijing, Seoul, and Shenzhen are also more expensive than in San Francisco. Add Tokyo and Taipei to the list of Pacific Rim cities with more expensive prices than Vancouver. This helps put Vancouver’s prices into context. Compared to most cities of the Pacific Rim, we’re still cheap. And lots of people are probably coming here with real estate money in their pockets from holdings they’ve sold (or in some cases held onto) back home.

Let’s run the same comparison checking in on income.

Price-Comparison-Numbeo-Oct-2018-B4

Compared to incomes, Vancouver stands out for its pricey real estate in North America. But again, in the broader context of the Pacific Rim from whence many of its immigrants arrive, Vancouver still looks cheap. Real estate is crazy expensive in Hong Kong and the major cities of Mainland China. It’s only slightly less expensive in Taipei and Seoul. Vancouver and Tokyo look quite similar.

The picture for rents is less dramatic than for purchase, and also holds different possible lessons. Average rents for available apartments are still crazy high in Hong Kong, Shanghai, Beijing, and Taipei, consuming over two-thirds the take-home pay of the average income earner. But rents aren’t far behind in Vancouver, San Francisco, and LA, which all hang close to ratios for Shenzhen and Guangzhou. There are a lot of high rent Pacific Rim cities. As I’ve argued before, rents are probably the most important thing to focus on in terms of insuring people can live in our cities. But it’s worth noting that available 1 BR rents take up under 40% of average incomes in Seoul and Tokyo. What might they be doing right in terms of taking care of renters that other Pacific Rim cities could emulate?

At any rate, as before, I’d love feedback on Numbeo numbers! They’re already showing up in academic papers. Do they look right to you? Way off? Better or worse than last time I checked in?

Hit me up with your thoughts!

Addendum: If you want to play around with my data download & the excel sheet I used for the above, here it is: Numbeo-Look-Oct-2018-B  Note that the income data from Numbeo was hand entered, because I couldn’t find a central source for it, unlike the pricing data by City.

 

 

 

Urbanism Axis & IMBY Allies

Last post I talked about how Vancouver’s election provided possible insights into the relationship between Urbanism (or IMBY-ism) as a political axis running perpendicular to more traditional Left-Right axes. Now we have results!

But first: a clarification. Initially I drew from the Cambie Report‘s clever crowd-sourcing of placements for political parties and prominent independents in Vancouver along both a municipal urbanism axis and social and economic left-right axes (ultimately combined), as follows.

Election-2018-positions1

Positions were allocated (and defined) by the wisdom of crowds. Not surprisingly, I received some pushback for accepting this wisdom – not everyone agrees with crowds! And that’s fair. Many policy positions and histories, especially within my area of housing, were actually more nuanced. Here I’ll provide a – still very rough – breakdown of how I see the axes providing important information about different positions and histories, which may be of use both for interpreting Vancouver and thinking through IMBY coalitions more broadly.

Election-2018-positions5

Starting with the Urbanist Right in the upper-right quadrant, we have a relatively familiar market urbanism: anti-zoning and libertarian inspired. Pro-housing everywhere “the market” wants it. Market urbanists tend to extol the virtues of density and disruption. YIMBY everywhere.

Moving clockwise, below we have the Preservationist Right. To add a bit of nuance, this is a position that I’ve argued actually much better characterizes the North American tradition: Rigid zoning for exclusive single-family neighbourhoods and more flexible market allocation of housing within a constrained urban core. Right leaning municipal coalitions offer a grand bargain between middle-class detached homeowners’ relatively conservative desires to be left alone and developers’ interests in making money downtown. The mantra goes something like: “Strong protections for me and the market for thee.” Or NIMBY in the Great House Reserve, YIMBY in the Urban Core.

Continuing around the clock to the Preservationist Left, we move toward the left-leaning reaction against the North American tradition. Anti-poverty alliances frequently identify developers as villains. This makes sense insofar as many alliances have borne repeated witness to the displacement that can result from unleashing market development upon the marginalized neighbourhoods of the urban core. Anti-developer politics can seem like a progressive end in their own right and can sometimes also win over middle-class voters (think “All neighbourhoods matter”). For a farther left subset, the socialization of housing seems the best bet for protecting those marginalized by the market. Lots more social housing is in order – but often concentrated in and meant to preserve neighbourhoods viewed as under development pressure. The orientation runs from NIMBY everywhere to PHIMBY (Public Housing in My Back Yard).

Moving up to the Urbanist Left, we find alliances that often view urban growth as good, both in terms of promoting diversity and in terms of reducing environmental impacts. Many accept that disruption is part of living in a city. But it shouldn’t be imposed unequally and policies should work to avoid displacement. Those I’ve also termed Inclusive Urbanists set their sights on returning exclusionary neighbourhoods to the urban fabric by reforming single-family zoning. They look to introduce social housing and diverse rental options to every neighbourhood. The tendency is YIMBY, but reform-oriented, with an egalitarian emphasis directed at diversifying single-family exclusionary hoods and large helpings of PHIMBY.

Ok, now let’s get back to what happened in Vancouver, where we had parties occupying each of these quadrants. Who won? And what does it tell us about IMBY-coalitions?

Let’s start with mayor.

Election-2018-positions6

The mayoral race ended up a showdown between two strong Urbanist Left candidates (Kennedy Stewart and Shauna Sylvester) and the strongest of the Preservationist Right (Ken Sim from the NPA – historically the epitome of the grand bargain party: NIMBY for detached home-owners, YIMBY in the urban core). Stewart beat Sim by less than a thousand votes (half a percentage point) in a real squeaker of a race.

Broadly speaking, both the Urbanist Left and the Preservationist Right fielded strong candidates. But what happened to the Urbanist Right and the Preservationist Left? The Urbanist Right candidate was clearly Hector Bremner, and at 5.7% of the vote, his Yes Vancouver party failed to attract much support beyond its passionate young base of market YIMBYs. The Preservationist Left initially had a party-supported candidate in COPE’s Patrick Condon (who early on made clear he wouldn’t run if the Green Party’s Adriane Carr had decided to try for the mayorship). But Condon dropped out after experiencing a stroke. Late in the race he endorsed independent candidate Sean Cassidy, who failed to attract much other support. Even adding in support for the fringe candidacy of IDEA Vancouver’s Connie Fogal (widow of progressive hero Harry Rankin), the Preservationist Left failed to crack 2% in the mayoral race.

So can we collapse the field, calling YIMBYism a left-wing phenomenon and NIMBYism mostly driven by more conservative impulses? Not quite so fast… let’s turn to Council! Here I note the average % of total council votes accorded to candidates in each party, highlighting the parties that actually won council seats (no independents won seats, despite overall respectable showings). I also provide the range of averages for major parties in each quadrant.

Election-2018-positions7

While the mayor’s race seemed to reduce relatively neatly to a singularly important Urbanist Left – Preservationist Right axis, the council race sees a real and strong split between the Preservationist Left (COPE and the Greens) and the Urbanist Left (OneCity and Vision), with the Preservationist Centre-Left ultimately receiving the most support. Indeed, the new council will be made up of five members from the Preservationist Right (NPA), four members from the Preservationist Centre-Left (three Greens and one COPE), and one member from the Urbanist Left (OneCity).

So what does this all this tell us about IMBY coalitions?

First: it’s important to distinguish those BYs: the backyards of Single-Family House neighbourhoods are treated differently from the backyards of the Urban Core.

Second: Right-leaning coalitions tend to do well in cities only when they leave the back yards of Single-Family House neighbourhoods alone. So far there’s little evidence that a right-leaning YIMBY coalition can win, though this could change in the future, as single-family neighbourhoods continue to lose population.

Third: Most YIMBYs lean left by quite a large margin, and left-leaning YIMBY coalitions can win. A Lefty Urbanist won the mayoral race in 2018, despite the competition from an exceptionally strong contender in the same quadrant. Moreover, Vision Vancouver’s coalition held power for the last ten years until their organizational implosion in 2018, and still placed well ahead of most other parties even if they won no seats.

Fourth: There’s a real and consequential split between Lefty Urbanists and Lefty Preservationists. I think this is often about perspective. From the point of view of anti-poverty activists working in the urban core, developers almost always look like villains (non-profit developers MAY be exceptions). From the point of view of people feeling excluded from cities’ vast tracts of single-family neighbourhoods, developers look like potential allies. On the flip side, the path to political success often runs through middle-class homeowners, and it’s easier to get them on your side by promising it won’t inconvenience them much than by suggesting they might need to sacrifice some parking or sunlight on their gardens. Vancouver’s Green Party, in particular, has walked this line to great success.

Theoretically, this election should put to rest the notion that all or even most YIMBYs are mostly market-oriented. After all, if they were they would’ve come out in mass for Yes Vancouver. Instead YIMBYs seemed to support Urbanist Left candidates in numbers easily surpassing support for the Urbanist Right. But to be fair, this was also a really messy election, witnessing the organizational implosion of the reigning party (Vision) and a confusing profusion of new parties. This likely benefitted those older organizations that managed to avoid imploding (Greens, COPE, and above all the NPA, who recovered strong from a shaky start), above and beyond informed platform comparisons.

Final Question: Can those Urbanist Leftys who made it into office this year work with their Preservationist Left or Preservationist Right colleagues? I’m guessing efforts to Make Room in single-family neighbourhoods are going to slow down again after Vision’s successful last-minute drive to introduce duplexes (with suites!), opening up all nearly all lots to four potential dwelling units (2x owned, 2x rented) across the City. Given housing plans put forward by the NPA and Greens, maybe we’ll eventually get matching legalization of an additional main unit rental suite (1x owned, 2x rented, 1x laneway rented), legalizing what’s already happening on the ground in many places. It’s less clear what will happen in the urban core, where alliances may shift project by project (remember, NPA councillors look pretty market YIMBY outside of single-family zones).

Let’s animate that GIF:

Election-2018-outcome-anim

 

Addendum: (preliminary) vote tallies from the City of Vancouver obtained here! Looking forward to the voting location breakdown we got from the 2014 election.

 

 

Fact-checking Vancouver’s Swamp Drainers

[co-authored with Jens von Bergmann and cross-posted with MountainMath]

Swampy facts: the dark, broken, and ugly side of housing talk in Vancouver.

Down south of the border, a politician who shall remain nameless campaigned on “draining the swamp” of Washington D.C., trafficked in countless conspiracies, and lied his way into office. His lies painted a picture of a United States turned dark, corrupt and menacing. He promised to fix it, Making American Great Again, mostly by shutting down globalization and kicking out the immigrants.

In Canada, we like to think we’re immune to this kind of rhetoric. But a strain has made its way into discussions concerning Vancouver, where the intersection of real estate, politics, and globalization are increasingly portrayed as a swamp in need of draining. We don’t believe most of those portraying Vancouver as swamp-like are intentionally lying (and in real life they surely favour the preservation of environmentally sensitive wetlands). Nevertheless many commenters are muddying the discourse with poorly sourced claims as a means of scoring political points and attacking various aspects of globalization.

It’s tricky to track down the spread of all the false claims out there. Fortunately a bunch of them were concentrated in a recent piece on “Dirty Money” in Macleans by Terry Glavin that views Vancouver as “a case study in the dark, broken and ugly side of globalization.” Recognizing that getting facts and interpretations right is often difficult for even the most well-intentioned, let’s work toward correcting a few misperceptions, line by line:

“At least 20,000 Vancouver homes are empty, and nobody’s really sure who owns them.”

Variations of similar statements permeate the media, with various degrees of factual accuracy. The most common misrepresentation is to refer to the 25k homes not “occupied by usual residents” as “empty”, which the above quote avoids by using an appropriately lower number.

The main issue with the above quote is that it’s portraying those “at least 20,000” homes as problematic vacancies, neglecting that that count includes moving vacancies around census day, empty suites (about 4000 of them), and units in buildings that completed around census time and did not have the time to fill in yet.

Accounting for these types of vacancies, we arrive at the ballpark of the Ecotagious Study based on BC Hydro data that found between 10,800 (for year-long vacancies) to around 13,500 (for four-month vacancies) and now the 8,481 empty homes through the Empty Homes Tax declarations, although some of those empty homes found via the EHT are outside of the universe Ecotagious reported on.

When quoting these numbers, the key question is what are the numbers supposed to be used for. If it’s to highlight “problematic” vacancies, then the Ecotagious numbers probably get us the best estimate for that point in time. Since then the number has likely dropped due to Empty Homes Tax pressure, we will have to wait until the repeat of the Ecotagious study to get confirmation on by how much.

And the reason we don’t know who owns them is not for some nefarious reason but simply because the methods we have for estimating empty homes (other than the ones caught by the Empty Homes Tax) do not allow for the identification of units.

“Another 25,000 residences are occupied by homeowners whose declared taxable household incomes are mysteriously lower than the amount they’re shelling out in property taxes, utilities and mortgage payments.”

That’s plain false, we have looked at this before. The 2016 census counted only 8,940 owner households with higher shelter costs than income. An additional 14,510 renter households paid more than their income in rent and utilities, making for a total of 23,450 households in the City of Vancouver that had higher shelter cost than income, most of which were renter households.

The wording of the sentence, followed by the next talking about tax avoidance in British Columbia real estate, seemingly suggests that the majority of these 23,450 households were cheating in some way. Let’s take a closer look at these households with shelter cost higher than income.

One of us (Jens) is partially responsible for bringing this stat into circulation and failing to provide more extensive context from the get-go.

figure-chunk-1

Looking more closely, we see that the bulk of these households are non-census-family households, probably roommates in many cases. Students likely account for a lot of the data. Single parents are also common. While there are some indications of irregularities in the data worth investigating further, broadly suggesting all these households are tax cheats is irresponsible.

“Non-residents own roughly $45 billion worth of Metro Vancouver’s residential properties, and non-residents picked up one in five condominiums sold in Metro Vancouver over the past three years.”

The first part is fairly accurate, CHCP reports that $43 billion worth of residential properties in Metro Vancouver were owned by non-residents. Of course that’s less than 5% of the total value of $884.5 billion.

The second part is a prime example of making statements without understanding the data. We don’t have data on non-resident buyers, presumably referring to buyers with primary resident outside of Canada at the time of the sale.

Considering similar statements in an earlier article by the same author, our best guess is that the author was referring to non-resident owners of condos that were built between 2016 and late 2017. Owners of recently built condos could be taken as a proxy for buyers if one makes some assumptions on resales.

Except the ratio of condo units built between 2016 and late 2017 that were held by non-resident owners is one in 7.1 for Metro Vancouver, and for the City of Vancouver that the previous article was referring to the ratio is one in 6.5. (CANSIM 33-10-0003)

In summary it seems the original statement is the product of playing loose with definitions, Metro vs City mixup and aggressive rounding to pump up the numbers.

“But Transparency International reckons about half of Vancouver’s west-side residences are owned by mystery trusts or shell companies.”

Big if true, a claim so outrageous that it needs data to back it up. It seems that this is based on a transparency international report that the author also referred to in a February column, where the author characterized this as “Transparency International estimates that perhaps half of Vancouver’s high-end residences are now owned by shell companies or trusts”. Now this has morphed into “about half of Vancouver’s west-side residences”. It’s good to remember what the Transparency International study actually did, it looked at the 100 most expensive properties in Metro Vancouver and found that 46 of these were owned by companies or trusts (not all of which have opaque ownership).

Via StatCan’s CHSP (CANSIM 39-10-0003) we now know that 5.61% of Metro Vancouver’s residential properties are owned by companies or trusts (or “non-individuals”), roughly in line with most other Canadian metropolitan areas in BC and ON as the following graph shows. Needless to say, the 100 most expensive properties on Vancouver’s west side are likely quite distinct from the rest.

Even after adding the non-resident owners to the non-individual owners, Vancouver still looks a lot like most other metro areas. In fact, the only metro area that really stands out is London, ON. Otherwise it’s the non-metropolitan portions of BC and ON that have the highest representation of company and trust ownership structures.

figure-chunk-2.company_non_resident

“In Metro Vancouver, homeownership costs amount to 87.8 per cent of a typical household’s income”

It does not. Most people spend far less, as the following graph on share of income spent by owners on shelter costs demonstrates.

figure-chunk-3

The author appears to be conflating running shelter costs of owner households with the RBC affordability metric which compares the cost of financing the typical home for sale in the region to the typical household income. The latter metric may (imperfectly) reflect some of the difficulty now facing those wishing to jump from renting to owning, but has little bearing on how much typical households currently spend in either category.

“Vancouver has also become a major global hub for organized crime networks based in China.”

Here the author immediately pivots to the opioid crisis and the suspicious transactions identified in the recent money laundering report concerning lax oversight of casinos, attempting to link these to broader affordability issues and to globalization. To be clear, both the opioid epidemic and money laundering are serious issues in their own right. The fentanyl crisis has killed way too many British Columbians. As a recent report by Sandy Garossino notes, the criminal organizations associated with money laundering through BC Casinos have also claimed multiple lives. We should be outraged by the crisis and the crime ring, but it’s wrong, as Garossino adds, that this, “mainly bugs us because we figure it’s driving up the cost of housing in Vancouver.” The opioid epidemic demands more sustained attention than it’s likely to receive as a prop for tarring globalization. That’s not at all what it’s about. It requires a comprehensive re-think of our health care systems, pain management strategies, and criminalization of drug use, and the biggest villain in the story so far appears to be a major American pharmaceutical company. As for money laundering, further reporting on its role within the real estate sector has been promised by the Attorney General, but so far it’s not clear that shady practices – while certainly present – have had much to do with driving up real estate prices. As multiple commenters have noted, even if all the $100 million so far reported to have been laundered in our casinos over 10 years was re-invested in real estate, it would represent at most tiny fraction of total real estate transactions. Property transfer tax data shows that Metro Vancouver averaged $5.2bn worth of residential real estate transactions each month in 2017, dropping to $4.4bn during the first 5 months of 2018. There are real reasons to be outraged over the opioid epidemic and money laundering. But the link between these issues and affordability remains tenuous, and insisting upon the link in the absence of further reporting diminishes the importance of the documented damage they’ve already generated without pointing toward any good solutions for affordability, the opioid crisis, or tackling money laundering.

“Freeland could have been describing Vancouver: ‘Median wages have been stagnating, jobs are becoming more precarious, pensions uncertain, housing, child care and education harder to afford.’”

This is plain false. To its credit, back in February the NDP government moved to make childcare much more affordable for British Columbians. Why ignore this progress? Moreover, Vancouver has seen strong jobs and income growth. To gauge wage growth, we look at full-time employment income for couple families, lone parent families and unattached individuals and compare the trajectories to Metro Toronto.

figure-chunk-4

We see that Vancouver CMA has overtaken Toronto for non-family individual income and lone-parent median income, and almost closed the cap on couple family income.

figure-chunk-5

This shows how Vancouver’s labour force participation rate has increased with respect to Toronto while the unemployment rate decreased. Lastly we can look at the regional job vacancy rate for the respective economic regions to see how Vancouver’s job market is much stronger than the labour force is able to fill.

figure-chunk-6

Afterword

In our hyper-polarized environment it is probably not enough to simply point out factual errors without further comment. So we take this opportunity to state that we strongly support stricter oversight and enforcement of money-laundering, as well as implementing measures to increase transparency in property ownership. We are also gravely concerned about Vancouver’s affordability problems. We’ve supported a number of housing policies recently put forward by local governments, including the empty homes tax, the school tax, and the boost to social housing investments, all aimed at fixing regulation and providing more housing to those most in need. It’s important to separate out what governments are doing right from where they might be failing. This is where swamp imagery fails us, blending everything together and dragging it all into the mud.

We think fixing our affordability problems is going to involve making tough choices and policy tradeoffs, and we should approach them with a clear sense of what’s at stake rather than mixed up facts, vague swamp-ish imagery and the sense it can all be blamed on the dark, corrupting forces of globalization. We’ve all seen where that last route can take us.

As usual (for Jens), the underlying R Notebook for this post that includes all the code for the graphs and numbers in this post is available on GitHub. Feel free to download it to reproduce the analysis or adapt it for your own purposes. Hopefully this kind of transparent and reproducible analysis can help establish a shared base of facts. And reduce the amount of guessing needed to make sense of people’s numbers and statements.

 

Update (July 27, 2018)

Several people have pointed out via Twitter and comments that inflation-adjusted income growth might be a better metric to use. And that’s a good point. In the context of housing we often have nominal housing prices in mind, so nominal income can be a good metric in this context. But inflation-adjusted incomes add another important perspective.

figure-chunk-Add

Here we used Canada-wide inflation estimates, Vancouver’s income growth looks even stronger when normalizing by CMA-specific CPI. (Digging into the reasons for this would probably make another interesting blog post.) The two graphs show the inflation-adjusted incomes, as well as change in adjusted incomes indexed to 2000. We can clearly see the growth in all categories, both in absolute terms, as well as in relative terms compared to Toronto. Despite this, the notion of “stagnant incomes” in Vancouver is quite pervasive in news stories.

How to Become a Resort City

I think many people are concerned that Vancouver is turning into a “resort city” – a playground for the rich – rather than a diverse and thriving city for all. But is this really happening? Yes. And I’m concerned too!

Certainly we see a lot of luxury cars and retail outlets, but we don’t really track wealth very well in Canada. Nevertheless owner-occupiers report their home values in the census and that’s where most wealth ultimately lies for Canadians. We can ask a simple question: where do the millionaires live? By millionaires I’m referring to anyone who reports owning and occupying a home worth a million dollars or more – not a perfect proxy, but not bad. The census uses self-reporting to get at this, and Statcan Table 98-400-X2016232 – in conjunction with total household numbers from Census metro area profiles – enables me to generate the following figure.

MillionaireHHs

Vancouver contains almost as many millionaires as Toronto, despite being less than half the size. Together the two metropolitan areas account for less than a fifth of Canada’s total number of households but over three-quarters of all owned dwellings worth on million dollars or more. To put matters differently, nearly 25% of Metro Vancouver’s households own a dwelling worth a million dollars or more, compared to just over 1% of the Rest of Canada (outside Toronto). We are literally concentrating the One Percent in terms of Canada’s wealthiest households.

Maybe Vancouver’s not quite a resort city yet, but it’s definitely in the neighbourhood!

So how did we get to this point?

In some ways Vancouver was predisposed to growth because it’s got lots of things people want: ocean, mountains, one of the mildest climates in Canada, a thriving port, a railroad line, lots of jobs, diverse ex-patriot communities, parks, strawberry-picking in the Agricultural Land Reserve, etc. This could explain growth overall. But everybody wants these things, not just rich people.

If it’s not just amenities that attract specifically rich people to Vancouver, then how are we becoming a resort city? Some people blame the fact that Vancouver’s amenities have been heavily marketed to rich people around the world in recent years. They have a point: this has certainly happened. But lots of other places try to market themselves to the rich as well. What makes Vancouver special?

We might not be attracting the wealthy so much as we’re systematically excluding everyone else who wants to live here. It’s about changing the composition of in-movers, so that wealthier and wealthier people tend to come (incidentally, this fits with research on neighbourhood change suggesting this is mostly how gentrification occurs).

How are we systematically excluding everyone who’s not rich? Easy! Under conditions of growth, all we have to do is preserve lots of urban land for millionaires and largely prevent anyone else from competing with them. Effectively this is what residential single-family (RS) zoning accomplishes in places like Vancouver, which we can see by comparing what proportion of detached and duplex housing is evaluated at over a million dollars.

ResortCityZoning

Across Metro Vancouver, nearly 60% of detached or duplex housing is evaluated at over one million dollars, yet the vast majority of urban residential land is zoned to support only these forms of housing. This is how you get a resort city, fit only for millionaires (with a little bit of room for their servant-tenants living in basement suites below – which is what duplexes mostly consist of in Canada). This compares to around of quarter of detached houses worth over a million in Toronto, and just over 2% in the Rest of Canada.

We can flip the question around to ask what percent of million-dollar dwellings are single-family detached or duplex dwellings. Strikingly, the answer in Metro Vancouver, Metro Toronto, and the Rest of Canada is pretty much the same: almost 90%.

Luxury zoning is almost entirely detached zoning. Across Metro Vancouver, the converse is also increasingly the case: detached zoning is becoming luxury zoning, affordable only to millionaires.

In places with lots of amenities – including jobs! – where people really want to move, growth is mostly limited by housing. If we only make housing for millionaires, we’ll increasingly have a city of millionaires. If we want to keep Metro Vancouver from becoming a Resort City, we’re going to have to tackle the zoning issue.

For this and more: yadda yadda yadda… book.

Thanks to Frances Bula, Jens von Bergmann, and Chad Skelton for inspiring today’s post and/or snatching away my afternoon! And if you’re interested in what you can do about reforming zoning, look into the platforms of municipal parties, like OneCity, that advocate for inclusivity across our urban landscapes.