Here’s uber-wonk Matt Yglesisas on housing affordability:[H]ousing can become more affordable if you build lots of new housing or else if your city becomes a less desirable place to live. Since few mayors or city council members want to outline “make the city a worse place to live” as a policy goal, that means you need to focus on allowing higher-density construction in desirable areas and on scrutinizing regulations that make new construction more expensive or more time consuming.
Overly simplistic, perhaps, but also importantly right. If population is growing and you want to keep a city both livable and affordable, there’s pretty much no alternative: over the long run, you’ve got to boost housing supply. (Well, you can also try monkeying around with what landlords are allowed to charge for housing—but rent control can have vicious unintended consequences both for livability and affordability.)
Of course, people who’ve sat through a day or two of Econ 101, or at least heard some talking heads on TV babble about markets, should understand that increasing supply tends to decrease prices. But for many folks the link between constrained housing supply and rising housing prices isn’t obvious—and I’ve got a few thoughts about why that is.
First: Many of the same people who talk about affordable housing actively oppose the construction of new housing. Some advocates for the poor equate new development with gentrification that obliterates existing low-cost units (which, of course, is sometimes true). Just so, many neighborhood activists pay lip service to maintaining affordability, but often dig in their heels to fight new development in their back yards. This contradiction—public figures who are outspokenly pro-housing affordability and anti-housing supply—adds to people’s confusion about whether new construction is the cause or merely the consequence of housing affordability problems.
Second: The word “affordability” conflates two separate problems: problems for the middle class, for whom supply and demand constrain the kinds of places people can afford to live; and problems for the very poor, for whom market dynamics are largely irrelevant. If you’ve got no money at all, boosting the supply of market-rate housing does nothing to help you find a place to live. Your problem is more about poverty than housing, yet we still talk about it as an “affordability” problem. Perhaps simply changing our terminology might affect how we think about the issues—or at least clarify which problem we’re talking about when we bemoan “affordability” trends in our cities. (Even for the very poor, of course, boosting the supply of housing is key to increasing access to housing; yet it’s often a process that happens through government intervention rather than the market.)
Third: The Econ 101 view of housing costs often falls short, or fails to jibe with what we see in the world. During the real estate bubble, for example, price trends had very little to do with fundamentals of supply and demand. (I’d argue that the bubble distorted everyone’s thinking about housing markets, since things were just so crazy for so long.) Just so, it’s easy to find neighborhoods or even whole cities where housing prices have gone up in tandem with supply—say, when neighborhoods improved rapidly or local economies boomed. The exceptions are so common, in fact, that it can be hard to see the rule. But over the long haul, and all else being equal, squelching new housing development tends to make housing costlier. Econ 101 doesn’t teach us everything we need to know about housing—not by a long shot—but we ignore it at our peril.
[Condo construction image from Flickr user edmontonenthusiast, released under a Creative Commons license.]
Bill Bradburd
Clark continues the oversimplification of this issue and in the process muddies up the waters further with false assertions. The first is to oversimplify the issue into a supply/demand argument. Housing affordability needs to be considered in two markets – purchasers (largely the middle class and up) and renters (largely middle class and down). While poverty and low wages create the conditions for the latter, all classes have housing needs. In Seattle there is a dramatic shortage of housing available to those at less than 70% of median income, and there is an oversupply of housing at 80% and above. Councilmember Licata held an affordable housing forum last week, and the panelists clearly debunked the idea that supply and demand factors in any significant way in this issue. Secondly, the amount of housing added in any year in Seattle has in recent history been about 1% – not enough to have any dramatic affect on the supply. Opening up the single family zones (which is often the coded back story) will not have any measurable affect on the problem and only creates further political upheaval.In theory developers should be leaping at the market demand for the lower demographic. The existing zoning envelope is capable of that. But developers and investors will respond to market demand in the rent/buy realms differently as will ‘consumers’. For example, even though mortgage rates are at an all time low, and there is nationally about a one year over-supply in new homes, no one is buying. Therefore, no is building either. So why aren’t they building apartments for the demand of the lower demographic. What Clark misses is the cost of construction and underlying land costs and how that affects production of housing that is affordable to below median income levels. Further compounding this is affordability for larger family units. A single person may be able to get by with 400 sq ft – a unit size that by its per square foot cost is more affordable. But a family of four will need at least twice that. This is why with a single wage earner for that family, housing in the city is out of reach.It is this reason why we have our dreaded sprawl. With land costs and building costs lower in the suburbs, that is where your affordable units are and this only contributes to daily commute congestion. Land prices in the city are higher, and construction costs are higher (and worsening as we make bigger buildings such as a multistory condos, and particularly as we get out of wood frame).Therefore just saying ‘build more’ ignores real development issues and how these factor into our problem.Secondly, the financial returns for developers are greater in producing units for sale, as well as having lower risks to them. Thirdly, just adding density externalizes costs of these units to the community and the city as a whole through greater demands on public facilities (water, sewer, amenities like parks and libraries) and poses potential threats to quality of life (noise, pollution, crime).Finally, up zoning as way to increase potential development only exacerbates the problem by markedly increasing land values and creating conditions that lead to further gentrification and displacement of people to points outside of the city.Because this argument is simply framed in the manner of the author, solutions to date have been ineffective and have tended to steer the city away from heading in the right direction. There are those of us that have been trying to ensure that as the city grows we do it in a manner that yields results we collectively desire – specifically those expressed in our Comprehensive Plan which is the standard for our adding density. While Clark derides those that are opposing the demolition of affordable housing stock, he ignores the fact that what is produced in its place is far from affordable to the extant community. His vague arm waving does not obscure this fact nor the ongoing gentrification. Nor does the simplified supply/demand argument solve the problem.What the city needs to acknowledge and address is merely adding density has no impact on the supply of housing affordable to a significant portion of our population. And that cities that have been through this stage of growth with unfettered development rules are now largely unaffordable to anyone but the wealthy. And, as an issue to Sightline, it is these conditions that lead to sprawl.What Seattle needs to do is to ensure that mechanisms it uses to increase density ensure that land values do not excessively escalate, that mechanisms exist to ensure that housing developers that target the less affluent have an equal shot to produce, to ensure that conditions are balanced so that for-sale units are no more favorable than rentals, and that all development pay its fair share and not continue saddle the city with a whole new list of demands and costs on our infrastructure. We also need to address the overall imbalance of jobs to housing in this city and in the region. As long as our daytime work population is hundreds of thousands of people greater than the nighttime population, we are further exacerbating our problem of transportation and transit costs. Housing density (and urban and cultural amenities) in the outlying areas needs to be increased as well (the New Urbanism) and we need to focus job growth in these areas as well to ensure balanced growth in the region. Seattle can not solve the problem alone.
Callie Jordan
I appreciate the longer and more in-depth discussion about housing issues. It reminded me of my (ongoing) efforts to understand the larger picture that would be impacted by the Henry George et al “land tax”. (I’ve been on the mailing list for Jeffrey Smith’s Geonomics for years and still can’t really quite get my head around it well enough to explain it to anyone else.)I can understand taxing the land itself rather than the improvements would encourage landowners to develop the land to its highest use, I’m not sure I understand how/if that would help the affordability. Here’s an excerpt from an article he submitted to the AIA, in which he uses “affordability” without really defining for whom:The biggest of the geonomic cities was Pittsburgh. For over 20 years, Pittsburgh levied a rate of 6:1 (land to improvement), which helped keep land prices low and enabled new construction. The city renewed its urban core without massive federal subsidy but by attracting private investment, and converted its most valuable location – where the three rivers meet – into a park without a grassroots effort to overcome developer resistance. Outside downtown, the new construction augmented the housing stock. Among major metro areas, Pittsburgh enjoyed the most affordable housing, which in turn yielded stable neighborhoods and low crime rates, also the best in the nation. Citing these factors, Rand-McNally named the Steel City “America’s Most Livable” for two years, 1985 and 1986. In 2000, the real estate lobby persuaded the city to revert to the conventional property tax. While construction starts have fallen across the state by about a percent, in Pittsburgh they’ve fallen almost 40%.He goes on to give examples around the world, where cities rejuvenated their urban core and/or had money to spend on parks or “citizen dividends”, but what did they do about the gentrification that replaces low income housing with higher value improvements (the essence of reducing “slum” conditions). Is this “land tax” a mechanism to increase density that will insure land values do not excessively escalate, as Bill points out we will need?? I would really like someone to read Jeffrey’s article and put it into context with the current discussion. I confess I get easily confused when things go past the Supply/Demand simplicity of Econ 101.
Clark Williams-Derry
Callie -Generally speaking, I think a land value tax—especially applied to already-developed areas—would likely generally housing supply. Here’s why.Under LVT, properties that currently sit underutilized—think, for example, of all the surface parking lots s in and around the downtowns of Seattle and Portland—would be taxed at the same rate as the apartment building or condo tower they’re next to. Facing sharply higher taxes for underutilized property, landowner look to develop the property (either directly, or by selling to others). If it’s zoned residential, then that means more apartments or condos. And while it’s certainly true that many people can’t afford to move into a brand-new housing unit—even if it’s not particularly upscale—it’s also true that expanding supply also tends to ease the upward pressure on rents. (With lots of housing in the market, people with lots of disposable income don’t wind up bidding up rent.)As I recall, people who’ve looked at the issue closely say that most property owners in a city like Seattle (and presumably Portland) come out ahead with this sort of tax system.Bill –In some places, I feel like you’re just looking for places to disagree with me. You say, for example –“What Clark misses is the cost of construction and underlying land costs and how that affects production of housing that is affordable to below median income levels. “But I completely agree with you!! (And so would any student of Econ 101.) Putting timing mismatches aside, rents are set at the marginal cost of producing a new unit of housing. When it becomes more expensive to produce a unit of housing, then rents go up. That’s one reason that the unit costs of multifamily housing tend to be lower than for single family zoning: land costs per unit are lower.
Michael Rouse
Continuing the discussion started by Allie Jordan and continued by Clark Williams-Derry, a land value tax would also directly (and positively) affect affordability. Here’s why.Assume that there is no land tax, and you have a $100K home on a $100K piece of property. If you have a 30 year mortgage at a 5% interest rate (and ignoring private mortgage insurance and property tax) you pay $1,073.64 per month. A full land value tax is equal to the interest-only payment on the $100K land value, so you are looking at a monthly payment of $536.82 (for the home) plus $5K/year, or $953.53 per month, a savings of $120.11 per month or over ten percent. The down side of this is that you are still paying $416.67 per month after the end of the mortgage, assuming the value of the land didn’t change. On the other hand, an LVT would allow for greater labor mobility, since many people change homes more than once every thirty years, and the lower cost of moving would help.