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.
Why do some buildings cost more to build than others? And how is it that some cities, as a whole, seem to build more cost effectively than others? Without getting into the specifics of how different markets work, I thought it would be valuable to outline some of the cost drivers that impact new developments. But keep in mind that this is by no means an exhaustive list. So, please feel free to add whatever I’ve missed to the comment section below.
Below-grade parking is hugely expensive. It’s almost always a loss leader. You lose money building it. That’s why parking ratios matter a great deal. If you’re building in the suburbs at 1 to 1 parking vs. 0.2 in the core, you’re simply building more of something that doesn’t make money. There’s also the question of whether you need to build a watertight below-grade, or if you can discharge any groundwater into the municipal infrastructure. Big cost difference.
Union vs. non-union construction labor.
Building stepbacks add cost and create additional complexity. To build more cost effectively, you really want repetition. But terraces are awesome. I get it.
Impact fees, development levies, and other government fees can vary widely across cities. As I’ve mentioned many times before on the blog, these line items can add up to 20-24% of the price of a new condominium here Toronto.
Time is expensive. One of the bigger line items in a development pro forma is financing interest charges. The longer things take, the more expensive the housing needs to be.
Markets are unique. Quebec, for example, has relatively low electricity rates. For this reason, it’s pretty common for homes in Quebec to use electric heating, which is usually pretty cost effective to install. According to this 2019 study, 68% of Ontario households rely on natural gas heating. In Quebec, the number is only 5%.
Depending on what you’re building next to, you may face additional costs. For example, if you’re building right up against a rail line, you may need to construct a “crash wall” in order to safeguard against possible derailments and you’ll probably need to up the STC rating on your windows because of the higher noise levels. These costs will not be insignificant.
In theory, land costs are supposed to be the residual claimant in a development pro forma. What that means is that you should back into your land value after calculating your projected revenue and considering all of your other development costs. If your revenue is lower, so too should be your land cost. Land cost as a percentage of total costs will, naturally, vary across different markets.
At a high level there are two components to the value of a house. There’s the value of the land and there’s the value of all the improvements. That is, the bricks, wood, and other stuff that form the actual house. When a media outlet runs a sensational headline about some shack in Toronto selling for, oh I don’t know, a million dollars, what it actually means is that the land in this particular area was just valued by somebody at this number. In fact, if the property is very clearly a “knock down” the improvements sitting on the land become a liability/cost rather than anything of value. Because whoever buys the land will almost certainly need to remove the improvements before they can build whatever it is they want to build.
This distinction between land and improvements is a valuable one for many reasons. Here’s one example. In cases where the improvements aren’t some shack, you may be faced with a scenario where a property can be valued in two different ways. You can value it based on the development potential of the underlying land or you can value it based on the income (either in-place or potential) that the improvements are generating, or could be generating with some hard work on your part. If the development value is greater than the value of the improvements, then there will be pressure to redevelop. Conversely, if the opposite is true, it is likely that not much will happen other than maybe capital expenditures applied to the existing building(s).
Of course, you could also run into a scenario where there’s little development potential and there’s zero ability to invest in the existing improvements, either because the market rents are too low in the area or because they’re capped and/or controlled in some way. In this scenario, it’s likely that not much will happen other than the normal and expected depreciation of the improvements. Maybe one day the development/investment math will work. But in the interim, you probably won’t be seeing any of those sensational media headlines.
Condo developers are merchant builders. They build a project and then move on. Because of this, there’s a belief that there’s little incentive to build for durability, in comparison to say purpose-built rental buildings where the developer might continue to own over an extended period of time. While it is true that putting on an operations hat will make you hyper-focused on everything from garbage collection to how you’re going to manage all of your suite keys, there are a few things to consider in this debate.
One, as developers we certainly think and care a lot about our brand and our reputation, both with our customers and with Tarion (warranty program). We ask ourselves: “What will our customers think if we do this?” Irrespective of the tenure we’re building, we want our projects to be carefully considered. And in the case of condominium projects, we would like our customers to feel excited and comfortable about buying in one of our future projects. That’s the goal. This is no different than any other product that you might buy that doesn’t come along with some sort of ongoing subscription.
Two, there’s often a spread between condominium and rental values. For example, let’s consider a brand new 550 square foot condominium in a central neighborhood of Toronto and let’s say it would cost you $1,300 psf to buy it today. (Obviously it could be more or it could be less depending on the area and the building.) Now let’s start with a rent and back into a value, using some basic assumptions.
Unit Size (SF)
550
Monthly Rent
$2,400
Rent PSF – Monthly
$4.36
Rent PSF – Annual
$52.36
NOI Margin
72%
NOI
$37.70
Exit Cap
3.75%
Value PSF
$1,005
Here I’m assuming that same suite would rent for $2,400 per month. I’m converting that to an annual PSF rent. And then I’m assuming that if you were managing a whole building of these kinds of units, your operating costs might be somewhere around 28%. Crude back-of-the-napkin math to get to a Net Operating Income (psf). Finally, I’m capping this NOI at 3.75%. We can debate my assumptions and if this were in a development pro forma you might “trend” the rents. But I find this comparison helpful. Here we are getting to a value of around $1,005 per square foot. Less than our $1,300 psf above.
The point is that the margins are tighter, which helps to explain why for a long time we saw very few purpose-built rentals being constructed in this city. So even though you might argue that the incentives are in place to build for durability, you do have to weigh that against the realities of what you can actually afford to build. Development is filled with all sorts of these tradeoffs. But if you and/or your investors really want a consistent yield, this strategy can work just fine. Personally, I’m a fan of the long-term approach.
Three, rent control policies can have an impact both on the feasibility of new projects and on people’s ability to actually perform maintenance. If you have a scenario where your operating costs — everything from taxes to utilities — are rising faster than your allowable rent increases, then you’re in a bad situation and you have zero incentive or financial ability to actually invest in the building, despite being a long-term owner.
Finally, there is nothing stopping a purpose-built rental developer from also being a merchant builder. i.e. Selling the entire rental building once it is done and it has been stabilized. So you could argue that we’re right back at my first point. Whether you’re selling to individual condominium owners or the entire building to one entity, you as the developer have to sit back and ask yourself: “What will our customer(s) think if we do this?”
Building on yesterday’s post about inclusionary zoning, below is a telling diagram from the Urban Land Institute showing which areas of Portland can support new development and which areas cannot. To create this map, ULI looked at achievable rents in each US census block to determine, quite simply, where rents will cover the cost of new development (all types of construction).
However, in their models they are also assuming a land value of $0. And typically people want you to pay them money when you buy their land. So in all likelihood, this map is overstating the amount of blue — that being land where new development is feasible.
But it does tell you something about developer margins. A lot of people seem to assume that the margins on new developments are so great that things like inclusionary zoning can simply be “absorbed” without impacting overall feasibility. The reality is that there are large swaths in most cities where development is never going to happen even if you were to start handing out free land.
This map is also helpful at illustrating some of the impacts of IZ. If you assume that rents are the highest in the center of the city and that they fall off as you move outward, then the outer edge of the above blue area is going to be where development is only marginally feasible. And so any new cost imposed on development would naturally start to uniformly eat away at the blue feasible area — that is, until rents rise enough to offset it.
Of course, this is a simplified mapping. Land usually costs money. Land values might also be highest in the center and fall off as you move outward, or there could be pockets of high-cost land. There may be more price elasticity in certain sub-markets compared to others. So the impacts of a new development cost may not play out as neatly as I outlined above.
Regardless, there will be impacts, which is why I find this map telling even if it isn’t fully accurate or up to date. Maybe some of you will as well.
A good friend of mine, who is also in the industry, once described real estate development as a three-legged stool. In order to develop, you really need three things: expertise, capital, and a site (i.e. land). This probably seems fairly obvious. I mean, you need to know what you’re doing, you need the money to do it, and then you actually need a place to build. But as simple and as obvious as this may seem, there are barriers to entry. Real estate is a capital intensive industry. And despite what the general public seems to believe about the pockets of developers, most are raising outside capital.
The thing about this three-legged stool is that you don’t necessarily need to have all of the legs at once, and in many cases you won’t. If you have two of them in place, it’s usually feasible to figure out and get the last one. For example, if you know what you’re doing (expertise) and you have a site (owned or “under control”), then presumably you have a development pro forma that makes some economic sense. And with those things, you generally should be able to find the capital that you need to execute on your project.
I’ve also met people who have managed to build this three-legged stool starting with only one leg. They didn’t have much development experience or capital connections, but they learned enough to figure out how to value development land. They then went out and started knocking on doors, eventually putting together a development assembly. They then took this assembly to developers (people with expertise) and the stool eventually got built. Starting with only one leg just means you’re going to have to work harder to fill in the others.
A one or two-legged stool won’t stay upright on its own. But hustle will hold it up temporarily while you figure out a creative way to attach the missing leg(s).
If you’ve bought land with the intention of developing it and you now think the value of that land has either gone up or down, there comes the question of what number you should plug into your development pro forma. Do you input what you paid for the land or do you input the current market value of the land? The former is probably more common than the latter, but in my view it’s important to consider both scenarios.
If the value of the land has gone up, it means that you think you could turn around and sell it for that price today. And that would mean you would be making a profit without doing anymore work and without taking on any additional risk. That’s an option that exists right here and right now (t = 0). What you want to get at in your pro forma, or at least understand, is the incremental profit margin from taking on the risk and brain damage of actually doing and completing the development project.
To do that, you need to consider the current market value of the land. That way you isolate your land margin from your build-out margin. The one problem with this approach is that the numbers may then tell you not to develop. In a hot market (which is not right now), it is not uncommon for land to get bid up beyond current fundamentals. There’s always someone else who is willing to be more aggressive.
In this case, you may find that most of the development margin is in the land. And you will start thinking to yourself, “How can anyone afford to pay this much? It doesn’t make sense.” This doesn’t necessarily mean that you shouldn’t develop. But at least it gives you a better understanding of the risk and reward trade-off that you’re about to take on. It might also tell you some things about the market.
If you’re trying to figure out how to make housing more affordable, it should be fairly obvious that it’s probably a good idea to actually understand the costs associated with building new housing. That is, more or less, the title of this recent series by Brookings about innovation in design and construction. The four-part series is based on the findings of a report that was written by Hannah Hoyt and published by Harvard’s Joint Center of Housing Studies and NeighborWorks America.
Now, costs vary by geography. Each city has its own nuances when it comes to development. And this should not be construed as a silver bullet. But what they are trying to do is identify design and construction savings to help the overall equation. Part of their argument is that building typology matters. Build smaller — hopefully out of wood — and you can bring your hard costs down. The problem with this thinking is that the trend lines are moving in the opposite direction.
Here is a chart from the same Brookings article:
In 2000, about 23%, or almost a quarter, of all multifamily units completed in the US were in a building with fewer than 10 units. As of 2018, that number had dropped to somewhere around 5%. At the same time, the number of completed units in buildings with 50 or more units has gone from 14% in 2000 to about 61% in 2018. Things got a little wonky after the global financial crisis, but generally the trend lines are pretty clear.
Some of this likely has to do with our “return to cities.” But I think the bigger part of this story is that development cost structures are pushing the market in this direction. For more on this topic, check out: Demystifying the development pro forma.
Yesterday I made a comment on Twitter about most people not understanding to what extent government bureaucracy inhibits the delivery of new housing in this city. It received a number of responses, including remarks about how development charges have also recently doubled and how this statement applies to pretty much every city out there. But there was also a comment about developers not being transparent and not properly explaining the impact to the public. In other words: please demystify the development pro forma. I thought that was a fair remark, and so this post is going to be a response to that comment.
Before I begin, it’s important to keep in mind that most developers have investors. These investors put up most of the money required for a project and in turn they take most of the profits. However, there is typically a “promote” in place, which is just an incentive structure that pays the developer more of the profits (disproportionate to the cash they invested in the project) if they perform and hit certain return benchmarks. All of this is to say that developers aren’t usually the ones holding all of the cash (which is what a lot of the public seems to think) and they are accountable to their investors to do what they said they would do.
Now let’s run through the costs that make up a “typical” development pro forma. For this example, I am going to assume that we’re talking about a 100,000 square foot mid-rise building; the kind that you might build and find along any one of Toronto’s Avenues. If we were doing this in real life, we would get more precise with the areas and consider gross construction area, gross floor area (city definition), and the net saleable/rentable areas. But to keep the math simple, we will ignore these differences. That’s the approach I’m going to take overall in the post. What you need to know, though, is that you have to pay to build the entire building, but you only get to collect revenue on a portion of it. That’s why the “efficiency” of a building matters.
Land
The value of development land is a function of what you can build and the revenue you can ultimately collect. So location matters a great deal. Based on the latest high-density land report from Bullpen and Batory, the average price of an unzoned mid-rise site in the City of Toronto is about $231 psf. So let’s assume a land cost for our project of $23.1 million. Assuming we can get land financing at 60% of the value of the land (loan-to-value), that would mean we’re putting up $9.24 million of cash (plus a loan guarantee!) and borrowing $13.86 million to start our project. At 5.25% per annum (interest-only loan), our annual interest charges would be about $727,650. From now on forward, we’re going to pay ~$60k in additional interest charges for every month that our project is delayed. Buckle up.
You should now begin to see why time is so valuable and why government bureaucracy can be so frustrating. As a developer, you’re heavily incentivized to move things forward, whereas it can often feel like everyone around you is trying to deliberately erect roadblocks in order to slow you down and make your project more expensive to build. Oftentimes, it is because it is less risky for them to punt things down the road and not make a decision. That is not the case for us and our project.
Hard Costs
Onto construction (or hard) costs. As many of you know, these have risen dramatically over the last 4 to 5 years. On some of our projects, we have added over $100 psf in hard costs alone. Part of this has to do with a busy construction market and part of this has to do with new building requirements: watertight undergrounds, new Green Standards, and so on. For our project, which is on the small side, let’s assume $360 psf for a total of $36 million. This would include our direct construction costs and our construction manager’s overhead (general conditions). We should also prepare for some of the trades to decline to bid on our project because it is too small and not worth their time.
Soft Costs
Soft costs include everything from consultant costs and interest charges to government levies and management fees. Like everything in your pro forma, these absolutely need to be broken out line by line. Don’t be lazy here. But for the purposes of this simplistic example, we’re going to use 75% of hard costs, which works out to be $27 million (or $270 psf). When I first started out in the development business, the rule of thumb was closer to 25% of hard costs. But times have changed. Government fees, alone, can make up about 1/4 of the price of a new condo in Toronto.
Adding up all of these costs, we’re at $861 psf or $86.1 million in costs. It’s now time to consider the revenue side. $1,000 psf seems like a nice round number, so let’s start there and assume we’re going to sell our condos for that. Typically in Toronto, the price you pay is inclusive of HST, so that liability will need to be deducted from our revenue line. It’s not a straight 13% because of the new home rebate, but the rebate also hasn’t been properly indexed since it was introduced and so the liability could still be upwards of 10%. (This is worthy of a separate blog post.) The result is $900 psf in revenue and a margin on costs that is less than 5%. No sensible developer would want to do this project. One misstep (or development charge increase) and you’re dead.
So let’s increase our condo prices to $1,100 psf. Maybe that will work. In doing that, we get to a margin on costs that is nearly 15%. Okay, now we’re in the range. But let’s say we just got delayed by 6 months (boom, interest charges) and our hard costs turned out to be off by $15. They’re actually working out to be $375 psf because of some new tariff and because the formworkers in the city are all tied up on bigger projects and couldn’t give a shit about our cute little infill project. Now we’re offside again in terms of our margin on costs. No problem, let’s try and push condo prices a bit more. Is $1,150 achievable? Perhaps. But ideally, given the above, we would want to be at $1,200 psf just to be safe.
This is an overly simplistic example of the math that goes into a development pro forma. But hopefully it begins to show you (1) just how many moving parts there are in a development project and (2) the kind of pricing that is required in today’s cost environment. Developers are reacting to the costs that they are being thrown and it is creating upward pressure on home prices. (See related post: Cost-plus pricing.) So far there has been enough elasticity in the market to absorb these price increases, but that may not always be the case. If you have questions about this post or disagree with any of my assumptions, feel free to leave a searing comment below.
There’s a narrative out there that all developers are uncreative and greedy, and if only they would start being more creative and generous, we could solve the housing affordability problem that is plaguing many (if not all) global cities. In other words, the solution to increasing the supply of low and middle incoming housing is simply a psychological reframing on the part of developers.
The problem with this mental model is that it ignores reality. Development happens on the margin. The market is competitive. It’s difficult to find developable sites. And it’s a challenge to make projects work. More often than not, you have to say no as a developer. No I can’t buy this land. No I can’t build housing here. And no the market will not support new office space here. Sorry, but no. (See cost-plus pricing.)
Development needs to give back. On the blog we usually call this city building. And that’s because it implies a greater sense of civic responsibility. Developers aren’t just building one-off buildings, they’re building a city. I believe wholeheartedly in this. But the belief that projects can be saddled with an endless array of government fees and civic contributions is a problematic one. There are limits — because markets have limits.
Many, or perhaps most, developers I know have a minimum project size that they will work on. That’s why you’ll hear people say, “No, that project is too small. I need at least X square feet or Y number of units.” Given that smaller scale development such as laneway housing and “the missing middle” are so in vogue today, I thought I would discuss some of the reasons why scale matters.
But first, it’s worth mentioning that “laneway suites,” as we have structured them here in Toronto, are intended to be built by individual homeowners and not by developers. The lots can’t be severed and most lots will yield less than 1,000 square feet. So this is a bit of a unique circumstance. As most of you know, I am a big supporter of this initiative.
When you get into larger developer-led projects, it’s a different ball game. For one, it’s hard to even find sites. And good luck if you need to deal with multiple owners as part of an assembly. Most landowners have pricing expectations that do not even remotely align with “missing middle” level densities.
But assuming you’ve been able to find land at a reasonable price, you still have to contend with the fact that projects have a lot of fixed costs, as well as diseconomies of scale. In other words, there are schedule, cost, and resourcing considerations that won’t change no matter how big or small you go. It’s still going to take this long and cost this much, and you’re still going to need a set of humans to manage it through.
This can then create a situation where there’s not enough margin for error. The project is simply too small to absorb any shocks, such as an unforeseen delay or an unforeseen groundwater concern that is now adding millions to your project budget. There’s a lot of risk with development and it’s prudent to have contingency room. That’s harder to do with smaller projects.
The other problem developers run into with smaller projects is that the construction subtrades also tend to think of them as smaller projects. They have their own set of fixed costs and margins to worry about. So unless you happen to catch them with an opening in their schedule, you run the risk of them telling you they’re too busy or them giving you a stinky price, which is just another way of them saying they don’t want the job.
On top of all this, there’s minimum project size inflation. If capital is not a constraint, there’s a tendency to want to do bigger projects (see above). And because the cost of everything keeps going up, it’s simultaneously getting harder and harder to make smaller projects pencil; unless you, maybe, go ultra luxury and ultra exclusive. But that’s kind of the opposite goal of this whole “missing middle” movement, is it not?
