Last week at the Better Boulder Happy Hour (B2H2) we tried to talk about affordable housing. The little nook at the Walnut Brewery was so packed that it was hard to even have a face-to-face conversation with folks, let alone do any kind of presentation that didn’t sound like an attempt at crowd control. Which is good I guess… but not exactly what we’d planned. I think a good chunk of the attendance was due to all the buzz generated by last Tuesday’s City Council meeting, and the talk of a citywide development moratorium. Anyway, it was a learning experience. We want these events to be informative, but also to get people talking to each other, and have it be more fun and social and network-building than a brown bag seminar or lecture that’s mostly going to appeal to the Usual Suspects, who are already engaged. We need to get more “normal” people to show up and engage on these issues.
In any case, Betsey Martens, director of Boulder Housing Partners (the city’s housing authority) got up and said a few words to the assembled crowd. She made a point which is in retrospect obvious, but that got me thinking anyway. The costs of creating additional housing in Boulder (or anywhere, really) can be divided up into three categories:
- Hard development costs — the cost of actually building the housing.
- Soft development costs — e.g. the financing and permitting costs, carrying costs associated with regulatory delay, organizational overhead, etc.
- The cost of land.
She pointed out that you can do all the work you want to reduce hard and soft development costs — using standardized designs, prefabricated buildings, streamlined permitting for affordable housing — but ultimately those optimizations just nibble around the edges of affordability. The real driver of housing costs in a desirable place is the cost of the land, which is pretty irreducible. If you’ve got a funding stream (as we do here from our inclusionary housing policy), then you can buy up a bunch of land and create housing on it, but there’s still an opportunity cost to be had for using the land inefficiently — the same money might have created more affordable housing.
The obvious way to attack this problem is to spread the fixed land cost across more dwelling units. You may not be able to reduce the price of the land, but you can share it with more people, decreasing per unit costs, and increasing density. Naysayers are quick to point out that all the density in Manhattan and Tokyo has not made them cheap. A common response is that they’re cheaper than they would have been if they hadn’t been more densely developed, but I’m not sure this is really the right answer (even if it’s true).
If the goal is to provide housing for people with a wide range of incomes in an economically efficient way, the problem with broadly allowing more intensive development — shared land costs — is that you can end up further increasing land values. If the increase in dwelling units is done by permitting the creation of more smaller units (rather than, say, increasing the allowed FAR) then I’d guess that the increase in land value is sub-linear (i.e. a hundred 40m2 apartments will rent for less than twice as much as fifty 80m2 apartments). This means there’s some real affordability being created — by virtue of giving people the option to buy less housing. But at the same time, upward price pressure has been exerted on land values, since the revenue stream that can be derived from the same underlying land has been increased.
If you do this once, it’s no big deal, but if there’s a general policy of increasing housing density to increase affordability by spreading fixed land costs over more people, then in the long run it’s not clear how much of that affordability endures, as land prices continue to increase, dragging housing prices up with them. Of course, not allowing people to share land costs doesn’t keep housing prices from going up either. You just end up with a smaller pool of housing being rationed out, that didn’t even get to take advantage of the initial intrinsic affordability that sharing land costs enables. (You also don’t get all the other benefits of increased density, like decreased per-capita infrastructure costs, energy use, and much reduced need for motor vehicles, which it turns out can contribute substantially to one’s overall cost of living). Then as long time home owners gradually sell off properties with enormous amounts of accumulated but as-of-yet unpriced equity, housing costs will continue to increase.
Boulder has a partial solution to this problem in deed restrictions and covenants that limit the appreciation of property values and the rents that can be charged for designated permanently affordable housing. Acceptance of these restrictions is a condition of receiving city funding. These restrictions also affect the potential financial upside for investment in affordable housing — and given a finite pool of capital to invest, most investors are going to want to do development that doesn’t have a limited upside. The result is that a lot of affordable housing is done by mission-oriented organizations: non-profits like the Boulder Housing Coalition, Thistle Community Housing, Habitat for Humanity, or quasi-governmental entities like BHP, or simply by homeowners who are willing to accept limited potential equity accumulation in exchange for knowing they will be able to continue living in Boulder long-term.
Is it possible to take advantage of land-cost sharing, without driving land prices up? Or equivalently, can you increase available land-use intensity without increasing the potential revenue streams that land generates? Would that be the result if increased density were only available to development that was entirely permanently affordable? What if only the additional density (above and beyond what would normally be allowed) were permanently affordable? How would you allocate the cost savings that resulted from being able to share the land across more people? It seems like as long as the revenues generated by the more intense affordable (or partly affordable) development were constrained to be no greater than a less intense market rate development, upward price pressure on land could be avoided.
Would it be possible for this type of development to be profit motivated? Could we allow developers to earn a return on the additional hard and soft development costs associated with creating more affordable units, but not on the land value? Or will providing the opportunity for increased profits in exchange for providing affordable housing inevitably result in upward pressure on land values? Is there a point at which that upward pressure on land values in the long term is sufficiently compensated for by the creation of new permanently affordable housing in the short term? Do we actually want to preclude people from taking advantage of the intrinsic affordability that results from sharing land costs if they aren’t interested in or don’t qualify for permanently affordable housing?
A lot of this seems to point toward non-profit housing development. If you’re a profit-maximizing developer and you have a choice between 50 market rate units, or 100 affordable units, and you know that you’re going to get the same revenue stream from each of them, what are the incentives for choosing the 100 unit option? Whereas if you’re a mission oriented developer (that’s trying to create affordable housing) clearly the 100 unit option is much better. Or maybe the profit maximizing developer could be compensated with decreased risk rather than increased return — streamlined permitting and thus reduced carrying costs and regulatory uncertainty for the 100 unit affordable option?
If you’re a mission oriented developer, what’s your source of capital, in the absence of a cash-in-lieu program like Boulder’s? Is there a role for profit driven but not profit maximizing developers? For community driven development like Germany’s baugruppen?
So many questions, but now it’s 2am, and I need to go lay on the deck, listen to the crickets, and gaze at the stars as I drift off to sleep.