The Liberals just announced that, if elected, they will form a new entity called Building Canada Homes (BCH) which will, "get the federal government back in the business of building homes." Broadly speaking, this new entity is proposed to have three key functions: it will build affordable housing at scale (including on public land), it will help to "catalyze" the private sector, and it will provide financing to affordable housing developers. There's a lot that is interesting in the policy teaser, but let's focus on function number one today: Do governments make good developers?

The outlined intent is that BCH will "act as a developer to build affordable housing" and "partner with builders for the construction phase of projects." So it sounds like they will not be constructors. The language they use also suggests that BCH will be an acquirer of land. Sometimes it will develop on already-owned public land, but in other cases it will go out and buy new land, sometimes offering it back to the market via land leases.

Acquiring new land will be challenge number one. As we have talked about many times before on this blog, land should be the residual claimant in a development pro forma. Meaning the value of land depends on what you can build on it. So if BCH is looking to build affordable housing and the rest of the market is looking to build some higher-and-better use, it will be very difficult for them to complete in the market. This is the same reason why, historically speaking, the City of Toronto has struggled to acquire new parkland with the funds it collects from developers. It can't compete.

On the flip side, it's very possible that in a downmarket, like the one we're in right now, BCH might be the only real buyer of development land. Affordable housing requires subsidies and if the subsidies BCH has access to result in both feasible projects and higher residual land values, well then they'll be able to win sites. But it will depend on the market conditions at the time. It also raises an important question: What is the right level of subsidy for the affordable housing that BCH intends to develop itself?

The second challenge is going to be execution. Development is a risky endeavor, but most of the time the private sector accepts these risks because they believe they will be compensated accordingly. And once they have taken on these risks, they become highly motivated to deliver for their investors and partners. Will the federal government be equally motivated? Perhaps. There are, of course, lots of examples of public housing developers in other parts of the world. But is it the most effective way to deliver new affordable housing? An alternative approach would be motivating the private sector to participate.

Development land, as we often talk about on this blog, should be the residual claimant in a pro forma. Meaning, start with your revenue, subtract your costs and required margin, and then see how much money is leftover to pay for the land. This is, in theory, how you should value land.

It's also the most disciplined way to go about your underwriting. In fact, it can be beneficial to not know the asking price or broker guidance for a new site until you've completed this exercise. That way you won't bias yourself.

However, in practice, it can be difficult to do all of this. In a rising market, you might find that there's always some other developer who is willing to be more aggressive on their assumptions, which means they will be willing to pay more for the same piece of land.

And so if you want to be in the game, you might find yourself doing the exact opposite: starting with the land price and then trying to figure out how to make the rest of your model work. We've all been there.

During this stage of the cycle, you get punished for being conservative and disciplined -- you don't win sites. But when the market turns, discipline and conservatism get rewarded handsomely. You then become thankful for the deals you didn't do. And I'm sure that many prudent risk managers are feeling this way right now.

It is very challenging to underwrite new sites today. Many of the assumptions that go into a pro forma are unclear and unknowable. And so the spread between what developer's models are telling them to pay and what landowners want to sell for is often significant. That is why everyone is trying to find "creative deal structures" that can be used to close this gap.

At some point, though, the gap will actually close; things will once again feel clear and knowable. I have absolutely no idea when that will happen, but I do know that when it does, it will then be too late from a maximum opportunity standpoint.

Because that's how risk works. Once the uncertainty is gone, it's no longer a risk. And if it's no longer a risk, then you're not going to be paid for bearing it.