Brandon Donnelly

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.

Subscribe

  • The 3-hour rule for high-speed rail

    I recently started following Adam Nathan’s posts about airlines, airports, and travel in general. This week he posted a comparison between taking the train and flying for 8 city pairs. The comparison looks at door-to-door time, factoring in the time it takes to travel to and from each station or airport; how long check-in, security and boarding take; and then how long you’re actually on board.

    What he found was that 3 hours on board a train tends to be the magic number. Once you pass this duration, travel times tend to flip in favor of flying. The one exception seems to be London-Paris, which is only a 2.5-hour train ride, but has a longer check-in time, presumably because you’re crossing a border in and out of the EU’s Schengen Area.

    I’m going to assume that the people planning Canada’s high-speed Alto train are similarly doing rigorous analyses such as these. Good case studies for us are Rome-Milan and Madrid-Barcelona. Both routes have on-board train times of about 3 hours and have a similar distance to Toronto-Montreal.

    If you want to ensure that high-speed rail is a success in Canada, make it faster, cheaper, and generally better than flying. This seems to include keeping the on-board time at or below 3 hours.

  • The math behind Toronto’s sixplex rollout

    Toronto started allowing sixplexes as-of-right across 9 wards last summer. Since then (and according to the Toronto Star), the city received 60 permit applications, more than half of which also included a garden or laneway suite (so 7 homes total). Our project at 571 Oakwood Avenue should be on the below map, but I don’t have access to the paywalled article to take a closer look:

    If we assume that 30 of the applications included 7 homes and the remaining 30 only had 6 homes, that’s a total of 390 homes, most of which were permitted as-of-right for the 16-month period between June 2025 and October 2026. Annualized, let’s call this 300 homes in the first year.

    What is obvious is that, for a city of 3 million people, this isn’t nearly enough. But we also know that it takes time for the market to respond to new policy changes and that we will certainly see an increase next year.

    In my view, if we saw a few thousand new homes delivered this way each year, it would be a policy success. So let’s say that we need to 10x the number of homes, which would mean bringing the annual project count from ~45 to 450.

    This feels doable, though it does remain to be seen how well this model will work in the inner suburbs. It’s worth noting that the one ward in Scarborough that currently permits sixplexes has yet to see an application according to this dataset.

    The big barrier is likely parking. Most of these infill projects are being built with none of it, and that means they’re only really feasible from a market perspective in areas that are walkable and have acceptable levels of transit.

    Even still, it’s not inconceivable to think that we could increase the annual application count to a few hundred. And this would be meaningful because it’s the most cost-effective way to deliver new infill housing. Single-family homes are obviously more expensive to build, and so too are higher-density concrete buildings.

    Add in six-storey buildings along the city’s major streets, and suddenly you’ve got missing-middle housing making a real contribution to Toronto’s housing needs.

  • The electrification of the global economy is here, but big bottlenecks remain

    The 31st UN Climate Change Conference (also known as COP31) is currently underway in Turkey. And one of the pledges being championed is a goal to increase electricity’s share of global energy consumption from about 23% to 35% by 2035 (the tagline is “35 by 35”).

    Of course, the electrification of the global economy is already underway. Here’s a fascinating chart from Bloomberg:

    Seeing what China has accomplished in 25 years is extraordinary. About a third of its total energy consumption is now electricity. Here in Toronto, I can also say that on all of our new infill rental projects, we are aiming to be 100% electric and have solar panels on the roof. We’ve also been exploring photovoltaic facades, and there’s a business case to be made for it.

    But the rubber does eventually meet the road somewhere, and the big challenge we are facing — one that I have blogged about before — is the high cost of new electrical services. It can be prohibitive on both small and large housing projects, where offers to connect can come with a bill that reaches into the high seven figures!

    The electrification of the global economy is the right direction, but it’s going to take a concerted effort across many different stakeholders.

  • A misleading condo narrative

    One of the market narratives at this point in the development cycle is that, as an industry, we’ve built too many “shoe box condos” and that what people really want are more family-sized homes.

    Now here’s some data from BC (via The Globe and Mail):

    As of the end of the third quarter of 2026, the Lower Mainland had 5,770 completed but unsold condos either in high-rise concrete buildings or low-rise wood-frame buildings. That total consists of 177 studio units, 88 junior one-bedroom units, 1,859 one-bedroom units, 2,865 two-bedroom units, 598 three-bedroom units, and 32 four-bedroom units. (The remaining 151 units could not be classified.)

    The region is also home to 1,253 completed unsold townhouses, the vast majority of which have two or more bedrooms.

    What this means is that over 66% of the standing inventory in the Lower Mainland is what one could call “family-sized housing” — in other words, they are homes with 2 or more bedrooms. This suggests that the prevailing narrative isn’t entirely accurate.

    The reality is that small condominiums and apartments exist because they are a natural market response to scarce land and high development cost structures.

    This is why simply mandating larger suites doesn’t solve the problem, nor does offering more “family-friendly” floor plan designs. Both ignore the more fundamental issue: large suites are expensive to build.

    I would also assume that some buyers are now substituting single-family houses for condos. I’m seeing that in Toronto. Whatever the exact situation, the answer isn’t as simple as too many small suites. In fact, based on the above figures, those are the ones that sold first.

    True market preferences will only reveal themselves when we lower the structural cost of building new housing and permit a variety of different building types.

  • Beyond left and right in city building

    I tweeted this morning (actually it was a post I scheduled weeks ago) that I generally don’t understand why so many city-building issues become political.

    The example I gave was corner stores, more specifically small-scale retail within neighbourhoods. Does wanting to be able to easily walk to amenities from my home place me on the left or the right of the political spectrum?

    I thought of this because every now and then someone will respond to one of my posts and sling some political moniker back at me in an attempt to debunk my argument.

    And each and every time it makes me stop and think: Am I really that? I don’t like the idea of being neatly placed in a particular bucket, and that bothers me far more than someone disagreeing with my point.

    When I ask AI to tell me what it thinks of me based on over 13 years of daily city-building posts, it says this:

    Based on your writing on brandondonnelly.com, you do not fit neatly into conventional partisan left/right buckets. Instead, your positions place you firmly in the pro-growth, market-urbanist / “Abundance Agenda” quadrant: culturally and socially progressive, strongly urbanist, yet grounded in market discipline, capital allocation, and regulatory pragmatism.

    This doesn’t sound wrong.

    But the result is that I often sit on different sides. I’m for bike lanes and neighbourhood retail, but I’m against inclusionary zoning and I don’t believe that rent control as a broad tool achieves its desired housing outcomes.

    At the same time, I also think you can look at specific issues from completely different perspectives.

    Bike lanes, for instance, are typically associated with centre-left progressives, but I see them as a capital-efficient, high-throughput form of mobility that increases individual choice and helps wean us off government-subsidized roads.

    Road pricing (or congestion charges) can similarly be viewed as a left-leaning tool that curbs emissions, promotes transit usage, and assuages the bike-lane crowd.

    But it’s actually the most fiscally conservative way to price roads. It’s classic Milton Friedman. Congestion is merely an economic failure: an artificial shortage caused by zero user cost on a scarce good.

    If I’m an advocate for anything on this blog, it is looking at each issue independently, trying to arrive at some understanding of the truth, and then being open to completely changing my mind if I end up being wrong.

  • The immediate returns of a car-free Oxford Street

    It’s only been about two weeks since London pedestrianized Oxford Street. While it’s certainly possible that more people are visiting the street purely because of its novelty, the initial reports are positive.

    In the first six days after it was pedestrianized, the broader West End area saw a 13% increase in transactions on Barclays debit and credit cards.

    What’s more, separate figures published by payments company Adyen show that in-store transactions across 40 different businesses on Oxford Street increased by 19% both week-over-week and year-over-year.

    While the datasets remain limited, all of this is promising and, in my view, expected. Oxford Street is coming off a difficult decade, during which vacancies rose to 13.7% in 2022 and average retail rents declined from £1,100 per square foot in 2016 to £600 during the pandemic.

    London is aiming, as it should, to compete with Fifth Avenue and the Champs-Élysées with this £150 million investment. This is an effective way to do that.

    Photo by Bryan Brittos on Unsplash

  • The end of CMHC’s multi-family monopoly

    The Department of Finance Canada just announced new mortgage loan insurance parameters aimed at expanding “missing middle” housing supply. The rules take effect for applications submitted on or after February 1, 2027, and they provide financing options for projects with 3 to 4 units and 5 to 50 units. Here’s the official announcement.

    What’s new and noteworthy is that the insurance rules are being amended to permit private mortgage insurers to back multi-family loans and to introduce greater flexibility for smaller three- and four-unit projects. In other words, the CMHC monopoly is ending for multi-family rental housing with 50 or fewer homes.

    At a high level, this is good. It will increase competition, because before, CMHC was the sole insurer, and it could often lead to problems of being stuck in the queue. I mean: We currently have a project that is shovel-ready with permits in hand, except we’re waiting on CMHC.

    The way I read these new rules is that they also create a new financing baseline. If you want to build new market-rate rental housing within the 5-to-50-unit bracket, you can expect an LTV of 90%, a maximum amortization of 50 years, a minimum DSCR of 1.1 (for 10-year terms), and a net worth requirement equal to at least 25% of the loan amount.

    If you want better terms than this and the project can support greater affordability, environmental, and/or accessibility features, well, then there’s still CMHC’s MLI Select points-based program.

    Streamlining and simplifying everything around missing middle housing is the key to unlocking more of it. Permitting, financing, design (things like single-stair), servicing (hi, Toronto Hydro), and environmental requirements (I’m thinking specifically of Records of Site Condition) all need to be fast and straightforward.

    Let’s hope this announcement improves the financing category.

  • Parkview Mountain House is now open for winter 2026/2027 bookings

    I’m in Park City right now working à distance, hanging with the family, and puttering around Parkview Mountain House in the evenings preparing for this winter season. I’m now excited to announce that PMH is officially open for winter 2026/2027 bookings! Here’s our website and here’s the direct link to our Airbnb listing.

    Now the important question: Will it be an epic winter with lots of snowpack?

    The honest answer is I don’t know, but the El Niño weather pattern in the Pacific is, as I understand it, a promising precondition. In fact, Ski Utah even launched a “Vote Super El Niño” campaign ahead of this year’s US elections. Maybe that was more just for fun.

    Regardless, there is actually a science behind Utah’s title of “The Greatest Snow on Earth.” I even bought a book on it, got deep into snow, and blogged about it here. Based on this cursory knowledge, I’m optimistic that this year is going to be a good one.

    Whenever I tell people about PMH, I’m often asked: Why Utah? And the obvious answer is the snow. But it’s also one of the most accessible ski towns. There are nearly a dozen resorts within an hour’s drive of SLC airport, and PMH itself is 27 minutes from the terminal.

    On this trip, we checked out Sundance Resort for the first time. Previously the Timp Haven resort, actor Robert Redford bought it in 1968 and renamed it the Sundance Mountain Resort, after his film Butch Cassidy and the Sundance Kid.

    I can’t speak to the mountain itself, but Bianca and I were both super impressed by the resort and the vibe. It has an elevated feel, and it’s clearly very popular throughout the year. We tried to have dinner there, but the restaurant was fully booked. So instead we bought a Sundance sweater for Vivienne. Now I can’t wait to snowboard there. (It’s a 50-minute drive from PMH.)

    If you like skiing and snowboarding or baby Sundance sweaters and want to visit Utah, you can check out Parkview Mountain House here.

    House photo by PMH creative resident Ryan Schwartz

  • When does the market wake up to underbuilding?

    In 2025, I wrote a post titled “Why 2029 will be the bottom for Toronto housing supply.” This was a fairly easy prediction to make, even four years out, because of how long it takes to build new housing.

    If the pre-sales and housing starts aren’t there (the leading indicators), then we know what supply will look like in the coming years. This remains true even if the market were to quickly rebound tomorrow; new supply simply can’t turn on that fast.

    But as clear as this future may be for those looking at the data, I don’t think the general market is looking toward 2029 and saying, “Damn, I better rent or buy now because in a few years housing is going to be scarcer.” And that’s because we’re still building and absorbing inventory from the end of the last cycle. That’s our current reality.

    If we assume 2029 will mark the supply bottom, the more important question for city builders is, “When will the market recognize this?”

    I have tended to be overly optimistic in my prognostications on this topic. At the beginning of this year, I said that things would feel much better in the second half of 2026 and into 2027. I’m not sure that’s true. Excess inventory is still being absorbed.

    At this point, I think we need at least another year for the market to really recognize that we’re underbuilding, which, by the way, is typically what happens at market bottoms. The hard part is nailing the timing.

  • SimCity 2000 comes to Toronto

    I just discovered this great isometric map of Toronto by John Fewell.

    He started with Google Maps using one fixed isometric view (30° down, aligned with the diagonal of the street grid), and then used Gemini to redraw each frame as an illustration. The overall map is intended to be a nod to SimCity 2000. (I loved that game in the 90s.)

    If you zoom in, you can even see cars driving around.

    At the time of writing this post, he had spent roughly $57 on Gemini and $21 on GPU rentals to produce it. Here’s more on his creative process. Imagine how much time and effort this would have taken before AI.

    Here’s a downtown screenshot:

    Here’s Junction House:

    And here’s me waiting for One Delisle to be rendered: