Affordable Housing is not Housing Affordability

Early 20th century houses in NW Portland.  Attribution

Amid fast-rising rents, affordable housing has come to the forefront of Portland politics as the necessary prescription.  But despite widespread political popularity, there is often confusion in the public about what affordable housing actually means. Affordable housing, as the term is used, means specifically subsidized housing, as opposed to market-rate housing which happens to be affordable. (I will use affordable housing here to mean subsidized housing and affordability to mean the general cost of housing) The federal government provides several programs to build and maintain subsidized housing, mostly administered through the Department of Housing and Urban Development.  Some states and local jurisdictions (such as Portland) also commit additional tax dollars to this purpose.  Some cities even require affordable units in new apartment buildings, effectively creating a subsidy by "taxing" the remaining tenants.  (Portland recently approved such a proposal)  Affordable housing is an important part of the social safety net that can advance a lot of our common goals: alleviating poverty, reducing income and racial segregation, promoting diverse neighborhoods, providing better social services to the very needy, improved mental health care, reduced homelessness and the list goes on.  There are lots of good reasons to support more affordable housing, and in the spirit of full disclosure, I work for an architecture firm that designs a lot of it.  

Rents in the Portland area have been growing significantly faster than incomes for the last few years. Attribution: Oregon, Metro

...But one thing that affordable housing can't do is reduce the cost of housing for the people who don't live in it.  By design, affordable housing is reserved for people with few economic resources, which by extension means those who are least likely to bid up the cost of housing.  In order to have an effect on the over-all cost of housing a large and ever-increasing portion of the population would have to live in subsidized units.  That's neither sustainable nor desirable.  In the city of Portland about seven percent of the population lives in subsidized units and that's well above average for an American city.  For the metro area as a whole the number is around four percent.  Even if we were to significantly increase those numbers - which would be a very ambitious (and worthwhile) feat - it would still leave the vast majority of us to deal with the spiraling costs of market-rate housing.  So it's important not to confuse affordable housing with housing affordability.  Affordable housing alleviates cost increases for a small segment of the population that is most disaffected by them - and that's a good thing - but it is not a solution to an affordability crisis. It is a much needed band-aid that, if we fail to take on the issue of housing affordability, will always be too small for the wound.

So what is a solution to declining affordability?  A real solution to an affordability crisis means addressing the problem at its root cause: it's too difficult and too expensive to build the additional housing that the growing city needs.  Despite all the cranes in the sky, new housing is not actually keeping pace with population growth.  Lots of people are moving to Portland and building housing to accommodate them is expensive, unpredictable and can take a long time.  That means less of it is built than is needed, which in turn drives up housing costs across the board.  It also means that units are smaller than they otherwise would be, which anyone who has toured a new apartment building has likely noticed. 

While Portland is experiencing an impressive growth spurt, it is hardly alone in this regard. Several cities have had growth spurts in the last few decades and not all of them have seen major increases in housing costs.  Cities that have mostly expanded at the periphery, where building housing is predictable, fast and inexpensive have generally seen relatively steady housing prices.  Houston, Dallas, Atlanta and Phoenix all fit that description.  Cities like San Francisco and Boston, that have constrained borders and a complicated regulations have seen exponential increases in housing costs.  (see chart below)  To be clear, I'm not advocating that Portland develop housing in the same way that Atlanta has - there are very real downsides to the extreme outward expansion of American cities.  But it is important to recognize that Portland's focus on infill development puts the city at an inherent disadvantage when it comes to controlling costs.  If the city wants to maintain its urban growth boundary and still keep housing affordable, it means going the extra mile to remove impediments to redevelopment.  

Circle size denotes percent population growth.  Note that population growth does not correlate with increasing housing costs.  Attribution: Issi Romem/Build Zoom

Few growing American cities have kept housing costs in check without aggressive outward expansion, but the same is not true internationally.  Tokyo has been dense and geographically constrained for a long time, and yet, in the city's recent growth spurt, redevelopment kept apace with population growth to such an extent that rents actually declined slightly.  Portland has also managed to solve the growth-affordability conundrum in its more distant past.  In the years following the Lewis and Clark exposition in 1905 the city grew at a rate that dwarfs its current growth spurt without creating a major housing crisis.  In part this resulted in the city pouring over the Willamette river creating the rambling east-side neighborhoods, but it is also evident in the northwest quarter of the city where apartment buildings rub shoulders with single-family mansions.   Michael Anderson has a very compelling recount of this period here.  The reason the city was able to grow so dramatically says Anderson is simple "Growing cities built and built and built — because until 1920 or so, there were no laws that said you couldn’t".   

There is a very real and well-documented correlation between barriers to constructing new housing and the cost of housing over-all.  A Portland where housing remains affordable to the middle class is a Portland where new development faces fewer barriers.  (My next post will look specifically at how we might do that.) The argument I've made here is basically the supply and demand argument many people remember from economics 101, but  I want to address few common objections to that argument because they tend to come from people who are genuinely concerned about the problem of rising rents and want to help.

The first is that housing is expensive because developers are greedy and "over-charging" for new housing.  I happen to know some developers that are very thoughtful, generous people, but even if they are exceptions to the rule, the point is moot.  If you make greed the cause of Portland's expensive housing you must also accept that the cause of Atlanta's relative affordability is its particularly generous developers.  Clearly this is not the case.  Greedy or not, developers operate in an economic ecosystem that defines what they can get away with.  In a market where new housing is easily produced, many people will produce it and any "greedy" developers will soon be forced to reduce their prices or be put out of business.  Conversely, the more complicated the land use process and the more expensive it is to produce new housing, the fewer who will be able to do it and hence there will be little competition to prevent developers from "over-charging".  In other words, don't blame the developer, blame their environment.

The second objection is that new housing is a cause of rising housing costs rather than a solution to it.  New housing is correlated with rising costs and generally new housing is more expensive (per square foot at least) than older housing, which sometimes leads people interested in helping low income people to oppose development.  Obviously this conflicts with the basic logic of supply and demand, but it's important to note that, As Noah Smith at Bloomberg points out, it's not a crazy notion.  Supply and demand often take counter-intuitive twists that don't match the x-shaped graph we know from our high school economics class.  Famously freeway construction has struggled with the phenomenon of induced demand in which expanding capacity leads to an even larger increase in demand for road space and thus more congestion.  (Congestion is a price paid in time rather than money.) The problem with the induced demand theory of housing is not the idea itself is flawed, it's that the evidence simply does not bear this out.  Looking at cities across time and comparing cities to one another shows that, in the big picture, fewer impediments to development, and hence more development, correlates to lower housing costs, not the other way around.  

Thus far the primary political response to decreasing affordability in Portland has been to propose more affordable housing.  This a good thing, but as I've said, it is not a solution.  Even worse, in the absence of a real discussion about over-all affordability, its way of changing the subject.  As an advocate for affordable housing I'm happy to see the issue getting so much press, but as a Portlander that values the economic diversity of the city I'm concerned we are not doing the things we need to do to keep the city affordable long term.  City policy has a major hand to play in determining how redevelopment responds to the need for more housing, and the extent to which the regulatory environment promotes housing growth is the extent to which the city remains affordable to the middle class.  The conversation about reducing the impediments to redevelopment is happening at the fringes through efforts like the Residential Infill Project, but it has been dwarfed by the focus on affordable housing.  This may be a sincere misunderstanding of the problem on the part of our city leaders, but tellingly, it is also a politically convenient one.  A real solution involves building a lot more housing - even more than we already are - and that involves some difficult conversations with people reluctant to see their neighborhood change.  Promoting redevelopment means taking on existing homeowners who have a financial interest in keeping housing expensive, and it means making some difficult decisions about the trade-offs between historic preservation and the benefits of new development.  None of these decisions are easy and none have a magic bullet solution where everyone gets their way.  They are real trade-offs, and as we know, trade-offs are never politically popular.  Politicians tend to promise that we can have it both ways even when we can't.  Unless there are advocates at the table forcing the difficult conversation politicians will continue to do what politicians do: change the subject.  

When it Comes to Housing, a Unit is not a Quantity

This post isn't a showcase of interesting buildings or neighborhoods, but an attempt to clarify a small linguistic error that often distorts our view of housing.  The word "unit" has become the standard moniker for dwellings in any "big picture" conversation.  Unit is an unfortunate and sterile word for the place that people make their lives, but the term has other flaws apart from its lack of intimacy.  The word unit has multiple meanings in our language, the most common of which is a consistent quantity of something; as in a unit of measurement. When we talk about housing though we use the word under one of its other definitions, not to mean a specific quantity, but to describe a distinct, individual part.  This is an important distinction because a unit of housing is not a quantity of housing.  Read that one more time: A unit of housing is not a quantity of housing.  A hundred five bedroom houses are not the equivalent of a hundred studio apartments.  They are not the same in either the size or in the number of people they are likely to house.  


This distinction is obvious when you think about it, but in the flurry of housing statistics that are thrown about it's often lost.  Units of housing are easily measured, while considering bedrooms and size to come up with some more accurate estimate of quantity is difficult.  But using units as a stand-in for quantity can lead to dramatic distortions.  To illustrate the point, Manhattan has far more housing units today than it did in 1900, whole districts full of high-rises have been built, but the residential population is roughly equal.  This is largely the consequence of moving from large households of many children, often with extended family members and boarders to small households with disproportionately low numbers of children.  On a per person basis, the large households of 1900 were simply a more space efficient living arrangement.  

Confusing units for quantity makes small units look like an attractive solution to a housing crisis (on paper) but making units smaller does not necessarily mean housing more people in less space.  Sometimes the opposite is true.  A family of four could live comfortably in a 1200 sf apartment, but a 300 sf studio pushes the boundaries of human comfort.  (Though it is certainly achievable)  And consider that in crowded cities, renting a room in a larger house or apartment is generally cheaper than renting a small apartment.  There are multiple variables at work here, but most significant is that it simply takes up less space. 

Household size is inter-related to the average unit size available in the housing stock.  It is not a one-way relationship, but a dynamic one.  Where I live in Portland, group houses of 20 and 30 something professionals are common as the city's average unit size is rather large while it has a disproportionate number of young, childless residents.  In a context of smaller units many of those group house residents would be inclined to find their own apartment.  Similarly where unit sizes are relatively large families are likely to take in relatives or boarders while smaller units will likely see more strictly nuclear families.  We don't get to rebuild the city from the ground up every generation, one way or another the people make a way to square their preferences with the stock of housing they inherit.  All of this is to say that when we talk about a shortage of housing we need to make some sophisticated judgements about housing quantity rather than just resorting to the simple but flawed method of counting units. 

Can Mid-Rise Buildings be Family Friendly?

The Modera Belmont, planned for inner east Portland is typical of new multi-family housing in the city.  It contains 202 units, mostly 1 bed rooms and studios.  Attribution: SERA Architects

 

Mid-rise, 5-over-1 buildings have become the dominant form of infill development in Portland and in many cities in the US.  It’s a building type that offers urban-scaled housing on a much lower budget than a concrete high-rise.  Very few of these buildings, however, are amenable to families.  In Portland, most newly constructed units are Studios or 1 bed rooms in the 500 square foot range.  While that reflects an influx of young, post-college professionals, the actual residents of these buildings include more families than you might expect, and at some point many more of those professionals are going have children.  The supply of single-family houses in most of the city is essentially capped by the lack of developable lots, so if the city is going to retain the new residents that have fueled Portland's renaissance we should be thinking about how we can create infill housing that supports multiple stages of life.  Mid-rise construction affords density and urban amenities, but in order to make it an attractive proposition for families we may need to rethink the prototype. 

So what would a mid-rise building look like if we truly designed it with families in mind?  Developers, thus far, have not had a lot of interest in this question, but one place it has been considered is in subsidized housing.

Nihonmachi Terrace, Northwest view (left) fourth floor plan (right).  Attribution: Pyatok Architecture + Urban Design

An interesting case study comes from Michael Pyatok, who has been at the forefront of affordable housing in the Bay Area for a generation.  I studied under Mike in grad school and, between his socialist rants, learned a tremendous amount about housing design. I’ve never met anyone who has such a firm grasp of the minutia of kitchen layouts and the exact closet space desired by different cultural groups. His approach to design involves a level of community participation that most architects would cringe at.  Beyond soliciting the residents’ advice on massing and preferred unit layouts, he also invites them to pick the style and suggest design elements.  It leads to a more colloquial aesthetic than most architects would appreciate, but it also leads to buildings uniquely fitted to a community.  If you want to design a family oriented building then working directly with the families that will occupy it is a very precise kind of market research.  Over time Pyatok’s firm has developed a variation on the ubiquitous 5-over-1 building that has become a template for their family oriented housing.  

Typical maisonette unit plans.  A variation of the three bedroom plan contains a "nested" bedroom configuration which allows the unit to be narrower and deeper.  Attribution: Pyatok Architecture + Urban Design

Nihonmachi Terrace in the International District of Seattle is typical of Pyatok’s work.  Most units in this project are two-story maisonettes, not flats.  Bedrooms are accommodated in an efficient upper floor plan while the entry level is an open living space.  According to Pyatok this plan is popular because it is considered the equivalent of a house.  It also allows every unit to have a living space with two exposures, limiting the bowling alley feel of many modern apartments.  However, this unit type also has a second advantage: it allows for a gallery access building without the persistent privacy concerns associated with a motel 6.  Bedroom windows are up out of the view of passers by while the kitchen forms a connection to the gallery.  Wrapping the building around a courtyard gives the gallery a sense of dignity and purpose that you won't find in a motel.  Pyatok laments the lack of landscaping budget, but even in its somewhat spartan form the courtyard provides a place at the center of the building that can be programmed or used as recreational space.  Since every unit connects to the courtyard directly, parents feel free to let their children use it as a play space.   A community room also opens onto the courtyard, reinforcing its role as the central recreational amenity. 

Nihonmachi Terrace, image of courtyard and gallery (left) and section (right).  The first four floors of this building are made up primarily of two story maisonettes with the gallery occurring only on entry level floors.  The  fifth floor is made up of flats intended for smaller households.  Attribution: Pyatok Architecture + Urban Design

 

Other buildings in Payatok’s portfolio are variations on the single-loaded courtyard type.  The Divine Legacy in Phoenix is an elongated version with more intensively developed landscaping.  Here the courtyard provides sheltered play space in a climate that is often too hot for outdoor activities. Flats are mixed with maisonette units to provide housing for a variety of lifestyles. 

Divine Legacy, courtyard view (left) and courtyard level plan (right).  Here the courtyard level is made up mostly of three bedroom flats with generous porches.  The floors above contain maisonette and loft type units accessed from a gallery.  Attribution: Pyatok Architecture + Urban Design

The Fox Court Apartments in Oakland are another variation still.  This plan divides the block into two courts, which Pyatok insists is necessary to prevent large groups of children from assembling and causing chaos.  The form resembles many late nineteenth century blocks in Europe, which contain several courtyards shared among multiple buildings.  In the European prototype blocks are often assembled by combining several single stair buildings with independent entrances.  Since these blocks contain a series of stair cores rather than corridors they can achieve through units that look out onto both the street and the courtyard.  American building codes require elevators and duplicate stairs the single-loaded, thus gallery access buildings are the only economical way to create through units.  In many European cities, 19th century perimeter blocks are a staple of middle class family housing.  Paris, Vienna, Berlin, Barcelona and a host of other cities that grew in the early industrial era are rife with examples.  Almost without exception these blocks contain a much higher concentration of families than can be found in most American cities.

Fox Court Apartments, East view (left) and courtyard level plan (right).  Here the north courtyard is shared with  a child care center providing convenient services for families in the building. Attribution: Pyatok Architecture + Urban Design

Mietskaserne Block typical of Berlin in the late 19th century (left) and rendering of Fox Court Apartments (left).  Both projects manage to create meaningful open space with a high lot coverage ratio.  Attribution: Charlie Gardner, Old Urbanist (left) Pyatok Architecture + Urban Design (right).

Pyatok's buildings are not the kind of high design that gets a lot of architects excited, but sexy modern design is skin deep.  It is fun to imagine what some of these buildings would look like with  lux finishes and beautifully constructed courtyards. (See my last post for more on that)  And the single-loaded courtyard building is certainly not the only building type capable of housing families in cities.  But this type is worth studying because from a construction point of view these buildings are a minor adjustment to an established formula.  They are still 6 stories tall with wood framing over a concrete deck.  From a development or finance point of view any new prototype requires a leap of faith.  Subsidized housing allows more flexibility to experiment with new prototypes that are not yet proven in the marketplace.  Developers, and more importantly their financial backers, want to build tomorrow what sold yesterday.  But those developers and bankers ought to be paying more attention to affordable housing if they want avoid the historic problem of doing market research by looking in the rear-view mirror.  After all, the future is not always like the recent past.  The millennial generation is heading towards prime child bearing years and a lot of them really hate the suburbs.  For cities its a housing challenge in the making and its not the kind of change we can adapt to overnight.  The development world and city planners need to start thinking about this issue now.  Some cities, notably Seattle and Toronto, have mandated a certain quota of family size units in new apartment buildings.  Its a good start, but ultimately creating housing en mass that works for a more diverse demographic requires both thoughtful city policy and prototypes with proven success in the market.  

 

A Look at Shared Courtyards

            As housing becomes denser, providing meaningful outdoor space becomes increasingly difficult.  Shared courtyards present a chance to create a valuable amenity with an economy of space.  The word “amenity” is important because it is development-speak for something that sells real estate.  The neighborhood swimming pool is an amenity, as is the golf course or the country club or the guarded entry gate.  These are high cost items that add substantial value to the real estate.  In an urban situation the space for amenities is extremely limited so there is an incentive to make more out of every square foot.  Courtyard’s can be a strategy for getting light and air to apartments, but they can also be important amenities that enhance quality of life.  They can provide play space for children where private yards are not an option, they can be a place to hold community events, a contemplative place to get away from family or roommates, a stunning view from ones apartment, and a place to cool down on a hot summer day.

A Beautifully remodeled courtyard in New Orleans, a city with a long tradition of urban courtyards.  Who wouldn't want to live here?

             I have been keeping a folder on my desktop for a few years, where I collect images of these types of spaces.  I am sharing a few of my favorites here.  The most successful examples are not only visually stunning, but are highly tuned to strike a balance between usable shared space and privacy for the surrounding units.  There is a similar balance of formality and informality as well as functionality and aesthetics.  When done right, courtyards are an amenity you can take to the bank.  A courtyard can be to a courtyard apartment building what a golf course is to a golf course community.  But with one critical difference, the golf course is an added extra, while courtyard usually something we have to build anyway.  

An everyday example

Another one from New Orleans

An informal courtyard in Spain.  Here the ground floor units have porches that act as a buffer between shared space and private space.

The courtyard leaves something to be desired, but the porches are great.

This one is from Savannah.  The entire surface is pavement yet the courtyard feels lush.  Modern courtyards are often built on top of parking making major plantings expensive.  Here is some evidence it can be done cheaply and done well.

A similar theme here, all pavement yet filled with green.  a

I have struggled to find good contemporary examples, but this one by Mithun is quite successful.  More pavement might have made this space more usable.

Some nice things are happening in this courtyard.  However the building looks like it sank into the mud about three feet.  Raising the ground floor units a few feet goes a long way towards achieving privacy.  Accessibility requirements in multifamily housing make this strategy difficult, but it can be done.  Raised patio's or porches might have made a much more exciting edge to this courtyard.  Unfortunately the developer chose to buffer the ground floor units with some dense plantings that, when they grow up, are going to block a lot of the light and views of these apartments.

This is a high end condo building in Manhattan.  There is a lot of thought and money put into this courtyard, but it still feels a little institutional to me.  Notice again the dense buffering of ground floor units.  Not the most elegant solution.

A nice start, but the architecture is not helping the courtyard here.

And finally, here is an old building with a new courtyard.  It feels a little stark with such young plantings and without any furniture, but no doubt it will mature with time and use.  

The Efficiency of Disorder ... An Interesting Case Study.

Chuck Mahron, who writes the Strong Towns blog has pointed out that efficiency and order are two very different concepts that are often confused.  He uses the traffic intersection as an example.  Streetlights, crosswalks, and clearly defined travel lanes all make the intersection orderly, but they don't necessarily mean anything about efficiency.  Sometimes the disorderly intersection, the one that forces people to slow down and rely on judgements and instinct can be more efficient.  Here is an interesting case study from the UK.  If you haven't seen it this video is worth watching.  

Vernacular Housing in Montreal ..The Three-Decker

Close up view of stairs, landings, and balconies, the vernacular language of

 the city.

Recently I went on a short trip to Montreal.  It is a beautiful city, and one of its charms is this type of vernacular housing.  It is a type of three-decker (stacked flats) that feels a lot like a row house.  My girlfriend and I decided to forego the hotel experience and try AirBNB. We got a room in one of these three-deckers in the Mile End neighborhood with a balcony overlooking the street.  Montreal is a low-rise city and even from the third floor we could see several miles to the church towers of the old town.  It was a wonderful experience.  I have never been on a trip where I spent a lot of time in the hotel room, but when your room is on a vibrant residential street and has a balcony with a view of the city, it changes the equation.  We enjoyed several bottles of wine and a few meals on the balcony watching life go by.  We watched the downtown fireworks on Candada Day, and watched commuters bike past over breakfast.  

Street view showing our room and balcony.

Mile End, along with many of the city’s inner neighborhoods, is made up almost entirely of three-deckers.  The neighborhood was developed extensively in the late nineteenth and early twentieth century as a very dense streetcar suburb.   Streets in this area of the city are strikingly regular with buildings built relentlessly to a setback line.  Cornice lines are set within a tolerance of a few feet and facades are flat. Rarely is a street lined with trees. There is a no-nonsense quality to this area of the city that is partly a product of the need to create affordable working-class housing and partly a reflection of the city’s French heritage.  Much like New Orleans, Montreal is more embracing of European style urbanity than its Anglo counterparts.  Having come to Montreal after spending several days in Toronto, this was especially apparent.  By comparison, Toronto’s classic row houses with their bay windows and elaborate gables seem a little silly, as if they are gyrating in place in an attempt to imitate a more suburban dwelling. 

Three-deckers showing a cut-in stair landing 

A typical three-decker with external stairs and balconies

Staying in a three-decker gave me an inside look at how these buildings work.  In many respects these apartments are more like a house than what most Americans are used to. Units have an independent entrance, light on several sides of the building, a generous amount of porch space and a small yard shared by the three apartments.  The anatomy of this type consists of three stacked flats, in an L-shaped arrangement that allows light and air into every room.  Each unit has a small balcony in the front and a larger rear porch that is set in the crux of the L.  The most interesting aspect of this type is the system of vertical circulation.  The ground floor apartment is accessed directly from a small stoop while an external stair leads to the second floor apartment and a door to an internal stair that connects to the third floor unit.  Each apartment has private outdoor space in the front of the building and in the summer time these spaces are used extensively.  On the ground floor this comes in the form of a terrace that occupies the six to eight foot setback while the second and third floor have a shallow balcony opening off the living room.  This arrangement affords an independent, street-facing entrance to each unit, and generates a unique aesthetic of steel balconies, landings and stairs set against masonry facades.  On more prestigious streets these elements are an elaborate display of craftwork. 

A contemporary row of three-deckers.  Some of the minor details have 

changed, but the type remains the same.

Montreal’s neighborhoods have continued to develop without fundamentally altering the three-decker type.  The modern examples I noticed have larger windows and presumably a more open floor plan, but in most other respects are no different from their predecessors. Probably the main reason this type has persisted is that it is cheap to build and uses expensive urban land effectively (nearly twice the density of equivalently sized row houses). Small buildings are often easier to accomplish where large streams of capital are not available.  A builder can put up a three-decker without incurring a lot of debt then sell it and repeat the process on the next lot. But as economical as these buildings are, they also add up to dense, attractive neighborhoods with a very unassuming scale.  

           

Three-deckers create an attractive streetscape and transition seamlessly 

with small commercial buildings.

About Me

I am an emerging professional in the field of architecture. I received my Masters from the University of Oregon in 2013 and have been in the profession since.  In my past lives I have been an artist, carpenter, furniture maker and avid observer of cities and architecture. Most recently my interests have centered on urbanism and new forms of urban housing.