Email to the city on proposed multiplex policy

With spreadsheets

An email I wrote to the city about the proposed multiplex policy. I had more to say than really fit into the survey (which I filled out as well).

Restricting the floor space to 1.0 FSR instead of 1.5 FSR means sacrificing on the order of $1M of value per project (roughly $400 per square foot x 2000-3000 square feet). That’s a huge amount. The city could capture most of that value as a density bonus payment, or accelerate redevelopment by leaving most of it to landowners, or anything in between.

[A roundup of previous posts:

• Why "gentle density" in residential neighbourhoods is so important - before the council vote in 2022

• Multiplex policy update from city staff - what the proposal says

• Fixed rezoning fees for smaller projects - using land-lift payments as a brake pedal

• Multiplex design concepts from Bryn Davidson - what plexes could look like

• Simple pro forma for a multiplex, with video. How economic viability depends on assumed selling price and costs.

• Why is the floor space limit for four-plexes and six-plexes so restrictive? A writeup for Reddit.]

---

To: multiplexes@vancouver.ca
Subject: Economic viability based on 1.0 FSR limit seems very sensitive to assumptions

Comments on proposed multiplex policy
Russil Wvong (morehousing.ca)

I’m writing to support the proposed multiplex policy. I’m really glad to see this moving forward. However, I’m concerned that the 1.0 FSR limit is too restrictive. If the assumptions about selling price or costs turn out to be too optimistic, few projects will be economically viable.

Economic viability as a bottleneck

We desperately need more housing. There’s four major bottlenecks: economic viability, rezoning, permitting, and construction. I would argue that construction should always be the bottleneck, and the city should make it a very high priority to open up the first three bottlenecks.

For multiplexes, no rezoning will be required, and the goal is that the permitting process will be comparable to that for a single-family house. However, the 1.0 FSR limit is not much more than the existing 0.86 FSR limit for a single-family house and laneway.

In order for redevelopment to happen, you need “land lift”: the value of the new building, minus the total cost to construct it, must be significantly greater than the current value of the property with its existing building.

A very simple model of what this might look like for a 33-foot east side lot, assuming a selling price of $1200 per square foot (maybe high) and total costs of $700 per square foot (maybe low). If the existing property value is $1.6 million, land lift is $400,000.

Land lift is volatile, because it’s a residual

The increase in land value required to motivate redevelopment is very sensitive to assumptions about the selling price and total construction cost per square foot. (In general, a residual value is volatile, just as the home equity in a home with an 80% mortgage goes up by 50% when the home price rises 10%.) There will always be a lot of uncertainty in the assumptions about what the future selling price and total construction cost per square foot will be. If these assumptions turn out to be off by even a small amount, this has an outsized impact on the land lift. In this  example, if total costs turn out to be $770 per square foot instead of $700, an increase of 10%, land lift drops by 70%. If they’re $800 per square foot or more, land lift is zero and the project doesn’t happen at all - it makes more sense to just leave the property as it is, with its existing building.

Given the uncertainty in estimates of future selling prices and costs, it’s very difficult for the city to exactly calibrate the restrictiveness of its policy to achieve a desired rate of redevelopment. And changing the policy is a slow process. This means that the city is always behind in its efforts to expand the housing supply. If the assumptions turn out to be wrong and the policy turns out to be too restrictive, it takes a long time for the city to go back and change the policy - by which time prices and costs will have changed again.

Use a high land-lift payment to limit redevelopment, instead of the FSR restriction

Instead, I’d argue that the city should set a less restrictive FSR limit, and use a high land-lift payment to limit the overall rate of redevelopment, say to 1000 sites per year. Thomas Davidoff suggests 1.5 FSR, and a land-lift payment sufficient to take all of the land lift (in this case, $1.4 million). If this payment turns out to be too high, it can be adjusted downward once or twice a year.

The key thing is that if costs rise and selling prices fall, bringing redevelopment to a halt, the city can simply lower the land-lift payment to compensate, rather than having to go back through the whole policy development process again. Having a larger potential land-lift to work with provides more headroom, so that the city can adjust policy quickly. (In this example, if costs turn out to be $800 per square foot instead of $700, the city has room to reduce the land lift payment from $1.4 million to $800,000, and it still has room to reduce it further if economic headwinds get worse.)

So with 1.0 FSR, you get 150 x 0.14 x 5000 = 100,000 additional square feet of space per year (assuming half are four-plexes on 4000 sf lots and half are six-plexes on 6000 sf lots), with up to 150 x $400,000 = $60 million in annual revenue to the city. With 1.5 FSR, assuming 1000 projects per year, you get 1000 x 0.64 x 5000 = 3.2 million additional square feet, or 32X as much, with up to 1000 x $1.4 million = $1.4 billion in revenue to the city. Given the desperate need for more housing, that seems a lot better.

What exactly are the tradeoffs required to have a less restrictive FSR limit?

From the public information sessions, it sounds like the Engineering department is thinking that with a modest increase in floor space and a modest number of projects, there won’t be too much additional strain on the sewer system, particularly from increased stormwater runoff.

To me this seems backwards - we have a desperate shortage of housing, and yet we’re trying to restrict it to match our aging sewers. I’d flip the question around: If the desired policy is 1.5 FSR, what is Engineering’s assessment of the resulting costs and impacts? For example, do sewer upgrade projects (already in progress) need to be accelerated in certain areas?

It could well be that the resulting assessment will make it clear that the cost of upgrading the sewers to keep up with 1000 redevelopments per year built to 1.5 FSR is prohibitive. But I would argue that the city shouldn’t be setting the limit to 1.0 FSR without knowing what that tradeoff is.

Infrastructure isn’t the only constraint motivating a limit of 1.0 FSR. But for other such constraints, the question is the same: What work would be required, and how much would it cost, in order to support a limit of 1.5 FSR?

Other comments

Some additional ideas for improving economic viability, keeping in mind that even small changes in cost can affect the land lift. Again, given the desperate shortage of housing, I’d suggest that the city should make it a very high priority to look for ways to improve economic viability.

• For multiplexes, don’t impose a “wedding-cake” constraint where the top floor can have no more than 50% of the floor space of the lower two floors.

• Don’t impose an on-site parking minimum. If the site is in a location where parking is desirable (e.g. it’s not close to public transit), then buyers will be willing to pay more for a unit with parking, and the builder can decide to include it.

• Allow a 50-foot lot to be split into two 25-foot lots, and allow zero-lot-line development. This would allow a six-plex on a 50-foot lot to be built as two triplexes, requiring only Part 9 of the building code (for small buildings) instead of Part 3.

Other options besides 1.5 FSR. Although they would reduce economic viability compared to 1.5 FSR, they would still be preferable to 1.0 FSR.

• 1.5 FSR corresponds to site coverage of 50% for a three-storey building. It allows an average unit size of 1500 square feet (minus unusable space, e.g. walls), which is a good size for a family.

• Bryn Davidson suggests 1.2 FSR (the limit for townhouses) would allow an average unit size of 1200 square feet.

• Another option: stick to the RS zone limit of 45% site coverage. For a three-storey building, this translates to 1.35 FSR.

• Or just set the FSR limit to 2.0 FSR and rely on the site coverage and height limits. With a height limit of three storeys and a site coverage limit of 45% or 50%, it’s impossible to get to 2.0 FSR. The effective FSR limit is then the site coverage limit x 3.

Miscellaneous:

• One aspect of the proposal is to reduce the FSR limit for a detached house from 0.7 FSR back to 0.6 FSR. While I understand the desire to create an incentive to build a new multiplex rather than a new detached house, I’m reluctant to endorse making the limits on floor space more restrictive than they currently are. Instead, I’d suggest introducing a density bonus payment for floor space above 0.6 FSR.

• Below-market ownership options: this seems difficult to administer in a fair way, and it also makes unwarranted assumptions about land lift remaining stable! I’d suggest sticking with a density bonus payment.

• Disclaimer - I’m not involved in real estate or development. I’m just extremely concerned about the increasing scarcity and cost of housing in Vancouver.

• There’s an interactive version of the spreadsheet linked from morehousing.ca/proforma.