Daily insights for city builders, delivered every morning at 6 AM ET. I’m Brandon Donnelly — a Toronto-based real estate developer and founder of Globizen. I’ve been writing here since 2013.
This discussion between Patrick O’Shaughnessy and Marc Andreessen is a great follow-up to my recent post about the productization of housing. Broadly speaking it’s about tech, software eating everything, and the future of the world. But if you skip to around the 15 minute mark, Marc talks about the growing divide in our economy between sectors that are changing rapidly and sectors that are changing slowly.
Examples of the former include things like computers, media, retail, cars, and a lot of the other stuff that we regular consume. Examples of the latter include things like healthcare, education, and housing (you know, the pillars of the American Dream).
The noteworthy problem with this divide is that the fast changing sectors are producing things that have been getting more affordable over time. The specific example that he gives is televisions. Think about how much more TV you can get today compared to when they were first introduced.
In contrast to this, things in the slow changing sectors keep getting more expensive. The same university education is exponentially more expensive today than it was a few decades ago, even though it’s far more important for people to have an education than to own TVs.
A similar thing can be said about housing. How much has really changed in terms of the way we build new homes?
One of the common threads across these slow change sectors, Marc argues, is strong government intervention. We restrict supply such that we can’t meet demand. We then respond to higher prices by trying to subsidize demand, but this only drives prices up even further. Because, at the end of the day, we haven’t addressed the underlying issue.
The result is a doom loop.
If you can’t see the embedded podcast above, click here.
Nabr, which I wrote about last year over here, recently announced its first residential project in San Jose’s SoFA district. Named SoFA One, the project is expected to have 125 apartments that will be offered up on a hybrid lease, own, and lease-to-own model. In this latter scenario, the company is saying that people will be able to buy with as little as 1% down. Construction isn’t scheduled to start until later this year, but if you’d like to get early access, you can add yourself to their waitlist, here.
As a reminder, Nabr is touting itself as a direct-to-consumer real estate company that aims to bring the same manufacturing and supply chain efficiencies that we have seen in virtually all other industries to the production of housing. This, of course, is not a new ambition. The flatlining of construction productivity is well documented, and lots of architects, builders, and entrepreneurs have tried to innovate in this space over the years. But it’s clearly a notoriously difficult problem to solve. So the obvious question here is: What is going to make Nabr any different?
Nabr is trying to productize housing. To do this, they’re building a vertically integrated process, going deep into supply chains, and trying to standardize their product offering as much possible. In the case of SoFA One, the base building is expected to consist of a CLT loft-style frame that can then be fitted out with various interior offerings. The idea here is that 90% of the build will be a repeatable system but that the remaining 10% is something that their customers will be able to customize — similar to when you’re buying a new car. The car is the same, but would you like black leather or brown leather?
Continuing with the car analogy, the company is also taking a move out of Tesla’s playbook for how they plan to roll out their products. The plan is to start at the top of the market (like what Tesla did with its expensive roadster) and then move downmarket as they drive efficiencies and cost savings in their delivery process. What they are trying to do is find the compounding innovation that has been present in most industries but that has been noticeably lacking from construction.
This all sounds great, but we know that buildings have a myriad of unique challenges compared to other products like cars and smartphones. My iPhone is the same as your iPhone, except for maybe the color and the case I put on it. But each development site is unique. Some have a high water table below it and some don’t. Some have adjacencies that will impact how you need to build and some don’t.
Each jurisdiction also has unique codes and regulations — everything from urban design guidelines to more or less stringent seismic requirements. Some cities have snow and some cities don’t. The list goes on. So what Nabr is going to have to do is create regionalized products with as much repetition as possible. And if they can generally lock the ~90% base building systems and just adjust the balance as needed, maybe that’s enough to do it.
At the end of the day, our industry is not completely void of innovation. It’s just a bit slow to change. We never used to build skyscrapers, but now we do. So I’ve decided to cast my developer cynicism aside. Today, we don’t have truly productized housing, but maybe we will.
As an aside, Nabr also recently shared their leaderboard of cities where people want to see a future Nabr building. Those cities are New York, London, Los Angeles, Toronto, and San Francisco.
Back in 2017, Portland, Oregon enacted new inclusionary zoning policies mandating that all new residential projects with 20 or more units must deliver a specified amount of affordable housing. Early accounts, by people like Joe Cortright of City Observatory, suggested that the market was reacting to this new requirement as you might expect. Developers rushed to get new applications onto the books and then there was a drop off in new housing supply.
Now that it’s been a couple more years, it is perhaps worth checking in on Portland. Cortright did that in the fall of last year and the housing numbers are continuing to fall. From 2019 to 2020, new multi-unit housing permits in Portland fell by more than 60%. I really don’t know the Portland market and so it’s hard for me to comment on whether it is solely the fault of IZ, but there was a peak in 2017 and now housing permits are down significantly. However, they were also down significantly during the financial crisis. It’ll of course be interesting to see how this plays out over a longer time horizon.
That said, a similar market response was recently reported in another Portland — Portland, Maine. In 2020, the city implemented a “Green New Deal” that stipulated, among other things, that all new residential developments with 10+ units would be subject to their new IZ policies. It has only been just over a year, but according to the city’s planning department, there were 756 new housing units on the books in 2020 prior to the new IZ policies. And since then, that figure has dropped to 139 new housing units. This is admittedly a small market and a relatively short time horizon, but it is still a data point.
As many of you know, I struggle with inclusionary zoning. Maybe it’s confirmation bias, but I just haven’t been able to find much data suggesting that it can meaningfully increase overall housing supply and the supply of new affordable units. So if any of you are aware of some good case studies outlining successful examples, please share them in the comment section below.
So 2021 was a pretty good year for condominiums here in the Greater Toronto Area. According to the latest data (Q4-2021) from Urbanation, this is what happened last year:
30,844 new condominium sales. This is a 69% increase compared to 2020, which saw 18,282 new unit sales.
Fourth quarter alone saw 8,361 unit sales, which is the best quarter on record according to Urbanation.
Unsold inventory dropped 26% year-over-year because sales exceeded the number of new project launches by over 4,000 units.
Average price for an unsold condominium unit in Q4-2021 reached $1,322 psf, which is an 18% increase compared to the year before.
Resale condominiums also did exceptionally well with 29,880 unit sales — a 49% annual increase.
All in all, some 60,000 condominium units were purchased last year in the GTA. Of course, some were ready to be lived in and some were future homes.
The Canada Mortgage and Housing Corporation (CMHC) recently published its latest data on housing starts, housing under construction, and housing completions. Here are a few of the highlights:
Canada saw 271k housing starts last year (2021). This includes single-detached housing and multiples, which captures semi-detached housing, row housing, and apartments (and other unit types). This is the highest number of annual housing starts that we have seen over the last five years. The range for the prior years has been between roughly 209-220k.
Ontario saw 100k (~37% of the country), Quebec saw 68k (~25% of the country), and British Columbia saw 48k (~18% of the country).
What I was curious about when I first saw these numbers was the split across the various housing types. Single-family homes, for instance, came in at 82k for all of Canada. So that’s about 30% of total housing starts. If you add in semi-detached and row, which I believe would also be all grade-related, you get to 124k or 46% of all housing starts.
Apartments and other unit types make up the balance at about 147k or 54% of all housing starts. This is kind of interesting because they now represent a majority.
Looking at Ontario, the percentage of apartments actually drops to 50%. But the numbers are much higher in both Quebec and BC at 69% and 63%, respectively. Again, this is kind of interesting.
Despite all of our deference to single-family housing, the numbers suggest that we are actually in the midst of building a different kind of country — one that entails people living in “apartments and other unit types.” Maybe it’s time we got more granular with this line item.
Note: CMHC defines “apartment and other unit types” to include not just apartments, but also stacked towns, duplexes, triplexes, double duplexes (whatever this is), and row duplexes. A number of these will, of course, be grade-related. But they still represent more intense forms of land use.
The “drive until you qualify” approach to finding housing that you can afford is a well established practice. Anecdotally, I can tell you that I have friends who are right now looking for a grade-related home under the C$1 million mark. This constraint, as most of you know, is pushing them to the outer reaches of Toronto’s suburbs. But if it were up to them, it would be their preference to stay in the city. According to the “two millennials” behind The Habistat, the average distance of an entry level detached house from the Toronto core (defined as a 3 bed, 1 bath under $800,000) is now 81.8km.
There’s a lot to be said about this. For one, home prices across many/most markets are way up. Earlier this week on the blog it was mentioned that the average price of a US home is up about 19% year-over-year. This is likely unsustainable. We are coming off of a period of easy money policies and at some point things will normalize along with the broader economy. Looking at the equity and crypto markets, it may be happening right now, but I don’t really know. (Fred Wilson wrote a post last year calling this “one of the great asset bubbles of modern times.”)
We know that the centralizing forces inherent to most cities have been weakened during this pandemic. For periods of time, they were completely off. So it is no surprise that we have seen greater decentralization (sprawl) than what might have ordinarily happened. I was in a (zoom) meeting this past week with somebody who has spent the last two years traveling around South America while working remotely. It sounded like a lot of fun and I was admittedly a little bit envious of her adventures. But as I argued at the beginning of this year, I think most people are going back to offices and this centralizing force will have an impact on real estate.
Because “driving until you qualify” is a function of an affordability constraint, it tells you certain things about consumer preference, but not all things. What I mean by this is that it tells you that somebody is willing to trade the cost of a commute for more space and/or the housing type of their choice. This has been an easier trade during COVID because the cost of commuting has been relatively — albeit temporarily — low for many people. So less of a discount for distance. But what I think this doesn’t tell you is what true consumer preference would be if all things were more equal and we increased housing supply and options in other areas of our cities.
At the same time, there’s a very real question of whether the measuring stick in the above chart should be a grade-related detached house? Is this a reasonable expectation in the same way it was for prior generations? I am not a fan of dictating what people should and shouldn’t do. But maybe 100km away from the core becomes untenable. And again, maybe if we increased both supply and options, we would find new housing preferences revealing themselves. I am specifically thinking of those who would prefer to stay in the city, but can’t find something they think is suitable.
At the end of the day, we can’t ignore the fact that we are profoundly hypocritical when it comes to the delivery of new housing. We acknowledge that we’re in a housing crisis and we acknowledge that we need more affordable housing (both for sale and for rent), and yet we continue to make it systematically more difficult and more expensive to deliver it. The development charges, parkland fees, and many other costs that continue to increase and get applied to new housing are a real worry to those in the industry.
It is a worry because we’re all wondering how much price elasticity is left in the market. That is, how much more can consumers afford before they stop buying and renting? It is a worry because it means that new rental housing, which has always been a challenge to pencil in our market, is now completely infeasible in many more submarkets. Our solution to all of this is to mandate a certain number of affordable units in new developments. But this is yet another tax on new housing.
To be fair, the delivery of new housing is subject to countless competing interests. This is arguably why it is such a tricky problem to solve and why there are no easy answers. But that’s what we do around here. We explore new ideas. And maybe, just maybe, there are other options besides just driving until you qualify. Next up (or soon up): A look at the competing interests behind new housing.
This recent article by Brookings is a good reminder of the all too important link between land use policies/patterns and GHG emissions. Because electric vehicles are cool and all, but they’re still not as efficient as just walking around and/or taking transit.
As has been argued before on this blog, we need to not only electrify our transport network, but we also need to change how we get around. And probably the best way to encourage a modal shift, is to plan and build our cities differently. Something that is simple, but not easy.
It also turns out that people who live in multi-family buildings tend to consume less energy (on a per capita basis) than those in single-family houses. So there are numerous benefits to encouraging denser housing on top of transit and within mixed-used communities.
With all of this in mind, here are some interesting charts from the above Brookings article.
This first one shows new housing permits in the metro areas of Atlanta, Chicago, and Washington DC, according to their urban, suburban, or exurban status. Here, Chicago is an outlier, with the “urban core” (defined as Cook County) now making up about half of all new housing.
If you look at the entire study period, the number is less. The urban core accounted for about one-third of new housing permits in Chicago, and only 15% of permits in Atlanta and DC. But in all cases, housing permits in the urban core have been increasing since the 2008 financial crisis.
But here’s the other thing. Looking at these next two charts, there appears to be a clear trendline toward more urban housing typologies. The first of these next two is showing single-family housing permits as a percentage of all new housing. And the second is structure type over time.
Atlanta is still building mostly single-family housing, but less of it. And based on these charts, Chicago has already passed its inflection point. DC is not far off. Every city region is of course going to be different, but it does look like there is some kind of broader housing shift underway.
Nice places to live — however you want to define that — tend to be expensive places to live. There are all sorts of reasons why this might be the case. Perhaps it’s on a body of water, next to a park, or it has some other redeeming qualities.
Daniel Herriges of Strong Towns makes a cogent argument, here, that when it comes to nice and desirable places it usually comes down to one thing: scarcity. Demand > supply. But on top of this, he argues that in most cases, the supply constraint is artificial.
Here’s an excerpt:
In fact, our shortage of nice places is almost totally self-imposed. And it’s precisely because 98% of the North American built environment is so blah that the 2% of places that are really well-designed environments quickly get bid up by the rich and become inaccessible to the rest of us. The solution to this isn’t to stop creating such places, but to create vastly more of them.
He goes on:
The same story applies to the countless row house neighborhoods of the Northeast, Chicago, and San Francisco. In city after city, the mass-market, working-class housing of its time has acquired a distinctly bourgeois reputation today. In all cases, the reason lies in economics, not design. What’s abundant becomes culturally coded as middlebrow; what’s scarce becomes culturally coded as elite.
We have talked before on the blog about how tastes change over time and how housing that was previously undesirable can sometimes/oftentimes become desirable given enough time.
My sense is that there are a number of factors at play here and it’s perhaps a bit difficult to decode where new “cultural coding” truly starts. But I very much appreciate Daniel’s scarcity argument. Scarcity drives so much in markets (just look at the NFT art market right now and the fixation on rarity tables).
But let me be the devil’s advocate. If we were to be successful at building no blah and all nice stuff, wouldn’t the rich just seek out a new 2% rarity? And if so, would the 98% still seem just as nice?
Either way, more nice places to live should always be the ambition.
In it are a few interesting figures about the housing market in the US. According to a recent survey that the company did, the average first-time buyer made 10 offers before successfully securing a home last year. The percentage of all-cash offers is also up to 25% from 15% a year ago. What is clear is that demand is currently outstripping supply. Based on these figures, housing supply in the US is at the lowest it has been since the early 1980s.
But of course, the real point of the year in review was to talk about all of the great things that Opendoor has been doing to digitize the real estate industry. Perhaps the most interesting is its focus on creating “all-in-one real estate transactions.” What this aims to do is consolidate the now separate processes of selling a home, buying a new home, and obtaining financing, into one digital workflow. Whether or not Opendoor is the one to do it, I believe that this is the future.
And what we have learned from other industries (that have successfully digitized) is that when you make something super easy, people end up doing a lot more of it.
Earlier this year, California joined Oregon to become the second state to pass policy that would allow additional housing density in single-family neighborhoods. Set to take effect on January 1, 2022, Senate Bill 9 requires that communities across California allow duplexes — and in some cases four units if they sever their lot — in most low-rise neighborhoods.
This is similar to what Toronto is looking at doing, though the details seem to be different. But as I mentioned before, sometimes you can have the broader permissions in place and yet very little building actually takes place because of other land use restrictions or market factors.
People in California seem to get this dynamic, because the Los Angeles Times just reported that some/many cities in California are now looking at local policies that would mitigate the effects of Senate Bill 9. In other words, they’re looking at policies that would make it harder to build the housing that this new law was hoping to unlock.
