As we have talked about many times before, the best answer to this question is that it's worth whatever money is left in your pro forma once you've accounted for everything else. This is what is called the "residual claimant" in a development model. And it means you start with your revenue, you deduct all project costs, including whatever profit you and your investors need to make in order to take on the risk of the development, and then whatever is left can go to pay for the land.
This is the most prudent way to value development land; but of course, in practice, it doesn't always work this way. In a bull market, the correct answer to my question might be, "whatever most market participants are willing to pay." And sometimes/oftentimes, this number will be greater than what your model is telling you, meaning you'll need to be more aggressive on your assumptions if you too want to participate. (Not development advice.)
Given that determining the value of land starts with revenue, one way to do a very crude gut check is to look at the relationship between land cost and revenue. This is sometimes called a land-to-revenue ratio. And historically, for new condominiums in Toronto, you wanted a ratio that was no greater than 10%. Meaning, if the most you could sell condominiums for was $1,000 psf, then the most you could afford to pay for land was $100 per buildable square foot.
However, this is, again, a very crude rule of thumb. I would say that it's only really interesting to look at this after the fact. Because in reality, things never work this cleanly. For one thing, there is always a cost floor. Don't, for example, think you can buy land in Toronto for $80 pbsf and sell condominiums for $800 psf, because this will not be enough to cover all of your costs. You will lose money.
Secondly, there are countless variables that have a huge impact on the value of development land. Things like a high required parking ratio, development charges and other city fees, inclusionary zoning, and so on. All of these items are real costs in a development model, and so they will need to be paid for somehow.
Typically this happens by way of higher revenues (in a rising market), a lower land cost (in a sinking market), or some combination of the two. But in all of these cases, it means your land-to-revenue ratio must come down to maintain project feasibility. This is why suburban development sites typically have a lower ratio -- too much loss-leading parking, among other things.
Of course, there are also instances where the correct answer could be a land-to-revenue ratio approaching zero, or even a negative number. In this latter case, it means your projected revenues aren't enough to cover all of your other costs, excluding land. For anyone to build, they will require some form of subsidy. And this is basically the case with every affordable housing project. They don't pencil on their own. (For a concrete example of this, look to the US and their Low-Income Housing Tax Credits.)
So once again, the moral of this story is that the best way to think about the value of development land is to think of it as "whatever money is left in the pro forma once you've accounted for everything else." Because sometimes there will be money there, and sometimes there won't be.
Photo by Jannes Glas on Unsplash
