#05 - Do Buildings Use Way Too Much Concrete?
Spoiler: the answer is "yes"
In This Issue
Do Buildings Use Way Too Much Concrete?
I recently had lunch with a structural engineer in Toronto. At one point I confided that I had a crazy theory, based purely on intuition after building a number of residential projects.
My theory: you could remove maybe 30% of the concrete in a typical Toronto building, and it would still be structurally sound.
Every market does construction in a different way. As an example, when I moved to Canada from the US, I was particularly surprised by the size and number of concrete shear walls in most residential projects. Similar projects back home would have few, or perhaps none at all! My lunch companion said this theory wasn’t crazy at all, and had a lot of fascinating things to say about this.
The idea that you could remove one third of the structure in a building might seem hard to believe if you haven’t worked in development and construction, but it’s something that industry veterans will know well. Twenty or thirty years ago, structural elements were more slender and had less steel reinforcement in them. Why would that be the case? It’s worth looking at how structural engineers design structural elements like columns, beams, and slabs.
Typically an architect will provide the engineer with some schematic drawings showing the general configuration of building elements: walls, floors, columns, stairs, and windows. After some back-and-forth coordination, the engineer can calculate the loads present on each structural element and design each one such that it can resist that load without cracking/breaking (called the Ultimate Limit State Design - ULS) or moving/deflecting too much (called the Serviceability Limit State Design - SLS). These loads include things like the occupants and furniture (“live loads” which change over time and are unevenly distributed), the structure itself and interior finishes (“dead loads” which are fixed), but also exterior forces like seismic activity, snow and wind. Some examples of design parameters are the size of a structural element, the configuration of steel reinforcement (aka “rebar”) inside the concrete, and the strength of the concrete mix.
If you knew the exact layout and intended use of a floor, including the number of occupants and the exact specifications of furniture and equipment, then you could calculate these loads very precisely, and then design a very efficient structure that would support those loads and little more. The problem with this approach is probably obvious to you. What happens if someone decides to install a new washroom or purchase new office equipment? You can’t have structural elements failing every time someone buys heavier furniture.
Safety Factors
To avoid this problem, safety factors are required for structural design. This means that structural elements must be designed with more strength than safety and serviceability require.
Most structures are designed to function with design live loads at stresses between 40-60% of their yield strength. Yield strength is the highest stress that a material can tolerate before permanent deformation happens. This 40-60% range is equivalent to a safety factor between 1.67x and 2.50x greater than the minimum required strength.
A large safety factor is a useful thing. It gives a structure extra strength in case interior/exterior loading changes, or to resist exceptional events related to weather, construction accidents, or other disasters.
In contemporary engineering practice, multiple overlapping safety factors are generally used: material factors, structural member factors, and load factors. For example, there are minimum design loads mandated in your local building code.
Here is an example of safety factor requirements taken from the Ontario Building Code. To use this table, an engineer calculates the anticipated loads (annotated as Dead, Live, Snow, Earthquake, and Wind below) and plugs them into the various cases shown. Each case provides a different combination of safety factors, and an engineer must design a structure to meet the highest one.
As you can see, the load safety factors range between 0.4 and 1.5, which are applied individually for each load type. This does not count the safety factors required for material strength or the dimensions of vertical/horizontal structural members.
On top of code-required safety factors, a structural engineer will typically add their own safety buffers. You might wonder why an engineer would do this, given how conservative the code is already. The professional engineers I asked put it plainly: a structural failure is a catastrophe for an engineering firm. It’s easier to over-design and virtually guarantee that there will never be a problem, rather than design to code-minimum and then face difficult questions if a problem arises later. This also makes the design process faster.
For example, let’s say there’s a building being designed with 100 columns. Many of these columns will have completely different design loads. An engineer could design each column individually. This would be very annoying for everyone involved, because it would take a long time and would lead to only minor differences between many of the 100 different column designs. The contractor and trades and inspectors would be very mad at the engineer for creating a bunch of extra work. For that reason, engineers tend to group columns with similar loads and then design them to be the same. Within each group, all columns need to be strong enough to meet the highest load. If a group has 10 columns, this might mean that nine of them are stronger than they need to be. So there’s a tradeoff between design efficiency, and the structural efficiency of the building.
But there are also risks from being too conservative. Let’s say the engineer chooses to optimize for an efficient design process and designs all 100 of these columns the same way — for the highest possible load. He makes all of them larger and puts more rebar in each one. You might assume that all columns will perform just fine, right? Not necessarily! Having a very large amount of rebar in a column makes it more brittle; this means that if it fails, it will fail more suddenly than a more elastic column would.
It’s worth noting here that the engineer has little control over the actual construction of their design. They can (and must) inspect construction progress, and they will review samples of concrete to ensure sufficient strength, but human and material error does happen. Any errors expose the engineer to liability, even when it’s not their fault. This pushes them toward more conservative design decisions.
So how much additional safety buffer does a typical office add on top of the code minimum? It varies tremendously. Some firms are known (and grumbled about) for being overly cautious, while others are more willing to pursue structural efficiency as might be required or requested by the client.
One engineer I spoke to said that when you add all of this together, you typically get structures which are 2.5-3.0x stronger than required for strength and safety. That means that buildings are probably using much more concrete and rebar than required. All of this extra concrete increases construction cost due to the extra material and labor. It also slightly reduces clear heights and the saleable/leasable floor area in a given project, so there’s a substantial hit to a project’s proforma when a structure is over-designed.
How much can this impact costs? One engineer I know recently completed a peer review where they saved a townhouse developer several hundred thousand dollars in rebar alone by questioning how the design engineer had grouped columns during the design process. If you can save six figures on rebar in low-rise housing, imagine what you could save in a large tower project!
Reasonable people will disagree about whether this is a good tradeoff. To a point, it makes sense to gain extra safety in exchange for some inefficiency. The question we need to ask is this: How inefficient are today’s structures compared to an “optimal” efficiency that leads to no structural failure, and a very low likelihood of serviceability problems?
A Natural Experiment
One engineer shared a truly incredible story with me.
An engineering firm was working on a high-rise mixed-use building. This kind of building will almost always have a number of transfer beams. In a tower project, higher floors will often have a different column layout than exists at ground level, or in a below-grade parking garage. This is because the dimensional requirements of apartments are different from those of retail space and/or parking spaces. A transfer beam allows you to transfer the loads from the structure above to a set of unaligned columns and walls below. Here’s a helpful diagram.
Here we can see an example where a column from the tower cannot continue to the foundation, so the weight of all the floors above the terminating column must be transferred to adjacent columns using a beam. This kind of transfer of forces is structurally inefficient, and requires much larger beam and usually more steel reinforcement. You can see the size of the transfer beam in the diagram is much larger than the other surrounding beams.
The engineer in our story had placed a standard beam in their drawings as a placeholder, assuming that someone would fully design the transfer beam once the overall project was more advanced. Amazingly, this never happened and the beam was never designed to account for the much greater strength needed to carry the weight from the floors above. The engineers missed it, the architect missed it, the city missed it, and the contractor missed it. Shop drawings were reviewed and approved. In any other project, this beam would have required hundreds of times more rebar and would have been much, much bigger. Only after the formwork was done and the rebar was placed did an engineer notice the problem during a last-minute site visit.
This engineer took this information back to their firm and reported the issue to their boss. The leadership team didn’t know how to react. And so… they didn’t react at all.
They ran the numbers again, and their calculations suggested that the beam would fail and crack. At the same time, they knew that if they acknowledged their error, they opened themselves up to liability and potentially losing a client. So they allowed construction to continue, and they quietly monitored the beam as the tower was built above.
And do you know what happened?
Nothing.
Even though this beam wasn’t designed to transfer the weight of 20+ storeys bearing down on it, there was no excessive deflection or cracking.
Now, I would argue that this engineer behaved in a highly unethical manner. Mistakes happen on every project, but a professional engineer under these circumstances has an obligation to stop construction and ensure the beam is properly designed. Setting aside the question of professional ethics, it’s an amazing demonstration that code-minimum design is so conservative that it can withstand forces many times greater than actually required. (There are some technical caveats here — certain types of oversight, like a discontinuous load path or a bad concrete batch not getting reported) could indeed lead to a catastrophic failure. Thankfully, that was not the case in this example.)
At the end of the day, I think it’s a fair to say that our buildings are amazingly safe, and that we could probably realize significant cost savings by making them slightly more efficient.
Business Updates
Imprint Development
In Issue #03 I wrote about the types of projects that I’m working on during Imprint’s first year. Nothing about this approach has changed. I am still primarily focused on new land assemblies and work-outs.
2023 was all about learning where the opportunities are, and going deep on planning policy (both existing and emerging) and where the opportunities are given all the uncertainty in the market. We managed to get two assemblies partially under contract, and are looking at more.
2024 will be all about solving problems for people. What do I mean?
This newsletter has been very helpful for connecting me with owners/investors/lenders who need help with their current projects. I’ve had lots of people reaching out to show me sites that need a complete overhaul, and I’m developing some creative deal structures that will help some of these projects avoid defaults and large financial losses.
There are lots of well-capitalized development firms who smell blood in the water, and who are waiting for great deals at steep discounts. I can’t compete with them when bidding on development sites; they can close faster and put up more cash. Instead, I’ve found a niche where I can offer more attractive terms by helping to save deals, and thereby avoid a fire sale in the first place.
Unfortunately, I can’t share much more detail about these projects (such as addresses, deal terms, or rationales for targeting specific neighborhoods) until we’ve advanced the projects a lot more. There are a few reasons for this, which I might write about in the future. But things are moving forward on both fronts, and I think this will be a good year for Imprint.
Urban Progress
One year ago, Urban Progress was launched as a non-partisan publication that explores issues related to urban growth and development. For the first time in history, a majority of the world's population now lives in cities. That means that the future of cities is the future of humankind, and Urban Progress is dedicated to envisioning what that future could look like.
I started UP while I was still working at Tricon Residential, purely as a passion project. So far, my co-founder and I have personally bankrolled the magazine as a proof-of-concept. We wanted to see if there was enough public appetite for its content (and enough people willing to financially support the venture) if we wanted to start a pro-growth, pro-immigration, pro-density media platform with an incisive approach to urban issues.
We do have a second goal for UP, which is a little more targeted. We are reasonably well-connected with industry, policymakers, and municipal staff in Toronto, so we have some insight into what policies and ideas are gaining traction among people who have the power to drive change. We are often surprised at what catches on — it’s often the least impactful, most expensive policy proposals that attract the most attention. One thing we often hear from insiders is that they want to do the right things, but don’t have the breadth or depth of expertise to come up with workable solutions, so they rely on industry and civil society. For better or worse, this has meant that industry groups and activist organizations have dominated the discourse for a long time. Each of these groups brings a valid, albeit narrow perspective.
We think there is a lot of low-hanging fruit when it comes to addressing our housing crisis. Things that have a great track record in other jurisdictions, some of which are highly technical and could be implemented without anyone expending serious political capital (single-stair egress, anyone?).
We want Urban Progress to serve as a catalogue of low-hanging fruit, which policy people can easily pluck for their own purposes.
Our current challenge is getting UP on sound financial footing. We don’t plan to make a penny on this venture, but we do have publication costs that we need to cover if this project is to continue. We have a number of subscribers on Patreon who generously support us. Unfortunately, it turns out that running a magazine is expensive! We are seeking a small number of donors who support our mission, and want to help position us for the next year of operations.
If you know anyone who might like to help, please send them my way.
In The News
Last month I included a short blurb about federal Housing Minister Sean Fraser’s letter to Mayor Chow, wherein he laid out a number of revisions that he wanted to see in a revised application for the City to receive Housing Accelerator funds.
City staff’s response to Minister Fraser was… a little disappointing. Here’s the formal response, which essentially says that the city was already doing everything requested in the Minister’s letter.
YIMBY groups and other housing advocates were watching this issue closely, and pounced to pressure Mayor Chow and Council to improve the city’s commitment to reforms.
Here’s an example from More Neighbours Toronto.
When the motion to approve the city’s revisions was brought to council, Councillor Brad Bradford moved to make some amendments, summarized below. This is all good stuff, but there are still no concrete commitments — mostly just Council asking staff to study several potential reforms which we already know are needed.
Nonetheless, on December 21st, 2023, the federal government announced that it would give Toronto nearly half a billion dollars from the Housing Accelerator Fund, so that suggests that the revisions were enough to satisfy the Minister’s team.
It’s not clear that these funds will move the needle on housing completions in Toronto. The HAF is set up to disburse funds to municipalities over a four-year period. It takes more than four years to get through the city’s planning process and complete construction. So it stands to reason that these funds cannot be meaningfully tied to housing completions that are unlocked by new policies that staff start studying in 2024. The city will have received most or all of the money before any new project is built, and before we have access to new data on completions.
Bottom line, it looks like the city has received a commitment of $471 million in exchange for promising to study potential reforms. Great for the city budget, not so great for addressing our housing crisis.
That’s a wrap on Issue #05.
My reason for writing this newsletter is to connect with more people, to learn what they think about my work, and to hear about interesting things that they are doing. So don’t hesitate to reach out.