Tuesday, August 23, 2016

Building affordable housing, where has the entry-level house gone?

One of the biggest topics of discussion in urbanist circles, and even beyond that, is certainly affordable housing. This debate is very politically fraught, even the term "affordable housing" is up for grab, with some people seeming to define it simply as "below-market price" housing.

Now, much has been written about the planning reforms that we need to achieve more affordable housing. But not much has been said about what affordable housing actually looks like. Sometimes, it feels like some people think that affordable housing is run-of-the-mill housing, just cheaper, and that's not how it works. You wouldn't expect a cheap KIA car to be equal in size, features and quality to a large Cadillac sedan... except for a third of the price.

So let's look at what affordable housing actually is, in a functioning market. OK, sure, any housing can be "affordable" to the resident if it's subsidized, but since housing is such an expensive spending post, subsidizing all housing just isn't really plausible.

Building affordable homes

In this era where planners love to build neighborhoods from scratch to a final state, believing that what is planned is what will be built and what will remain there forever, the idea of adding affordable housing is often first conceived as the idea of building housing that is affordable from the get-go.

Now, theoretically, any housing can be "affordable" if subsidized, either directly or by regulation that forces the price of some units to be maintained lower than market value. But let's ignore that possibility here, let's look at what affordable housing can really be. Without subsidies, developers have to recoup the cost of constructing homes from those who buy the homes, so a look at the actual cost of building a house should provide a glimpse into what affordable housing can be.

Like any project, the first step is to establish a budget. For simplicity's sake, I will suppose that an affordable housing option is 3 times gross household earning, given my sources, all the amounts will be in Canadian dollars, adjust accordingly for American figures.

OK, so in Canada, the median household income is 69 000$. The low-income threshold for an household of 2 people is around 25 000$. So, as a gross estimate, an affordable median housing unit should be about 210 000$, and an affordable housing option for a poor household should be 75 000$.

So, what does that buy you?

I'll use a catalog of housing plans in Québec which provides estimated construction costs for their different plans to find out. You can check out the website at http://www.planimage.com/fr/. First, I'll restrict myself to single-family structures.

So, a 210 000$-house would be something like this 3-bedroom 1 260-square-foot bungalow, with an unfinished basement that offers the possibility of doubling the living space (so total potential living space of 2 520 square feet).

Or a 3-bedroom 2-story house with 1 546 square feet of living space with an unfinished basement of 773 square feet, for a total potential living space of 2 319 square feet.

On the other hand, what does 75 000$ buy you? Well, looking at the catalog, houses don't even go that low, the most affordable house model is a 2-story 910-square-foot house with 3 small bedrooms and an unfinished basement that can increase total living space to about 1 300 square feet.

If we look at bungalows only, then the cheapest bungalow model is a bit pricier and a bit smaller (768 square feet), but with a potential of over 1 500 square feet after the basement is finished.

But all is not lost, because filed under "chalet" or "secondary homes" meant to be only occasionally used homes in the countryside, we can find the following.


These are different from the bungalows above in that they are lacking in basements, meant to be built on concrete slabs. Which means that these do not have expandable living spaces. Note that unlike the previous homes, where the construction cost per square foot was around 100$, the small houses actually have higher construction costs per square foot, around 150$.

Still, this shows that it is theoretically possible to build affordable housing in North America, even in the form of single-family houses, housing that can be affordable even for people at the threshold of poverty. And indeed, such small houses can sometimes be seen in Japan in clusters that seem to indicate public housing developments.

Probable public housing developments in Hokkaido, Japan, note the number on the buildings, typical of Japanese public housing developments

Small 1-story houses in Obihiro

Of course, there is always the option of multi-family developments, and the same site provides a range of low-rise options for multi-family housing.

This triplex offers three 2-bedroom 950-sf units for about 300 000$ (around 100$/sf)


This 4-plex offers four 2-bedroom units, each 910 square feet, for 383 000$, or about 100$ per square foot too.


This 6-plex offers six 2-bedroom units each about 1 000 square-foot big for 604 000$, or about 100$ per square foot.

All of these are low-rise, walk-up buildings, all offer similar construction costs per square foot to the house models. So theoretically, it is certainly possible to build affordable housing for the middle-class and even the poor. Caveat: construction costs do vary from place to place, the biggest variable being the price of labor, as most of a house's building cost is labor, not materials.

Note that all these buildings have common characteristics of being low-rise, of being wooden-framed structures, of having little public areas (for multifamily units) and lacking elevators or any complex mechanical system. From what I've read, this is the most affordable type of housing one can build (at least per square foot). As long as you satisfy these criteria and do not have a tiny building, the construction cost per square foot tends to be roughly the same, around 100$ per square foot.

On the other hand, high-rise constructions that require concrete frames, elevators, sprinklers, plenty of mechanical fan ventilation of inner areas and units, etc... tend, from what I've read, to cost 50 to 100% more to build than the low-rise units. Not only that, but these mechanical systems need to be maintained, which adds to the cost over time. So if an household can afford only a 700 square-foot apartment in a low-rise unit, maybe it could afford only a 350 to 450 square-foot unit in an high-rise.

I know, I know, a building's cost is not limited to construction cost. So let's talk about this.

Land cost

You can't just build housing in the air, you have to acquire a plot of land first. Since land is not built, but rather pre-owned, it's all a matter of supply and demand, without construction costs as a yard stick to estimate value in a functional market. However, not all land is worth the same, the desirability of the location varies and creates submarkets. For example, if you have subway lines in a city, land within walking distance of a subway station may well be a submarket for land, and that type of land may well be in shortage even if there's plenty of undeveloped land waiting for buyers all around the city, land that just happens to not be near subways.

Furthermore, a building should have to pay for the municipal infrastructure needed to link it to a city's street grid and public services (drinking water, sewers). I can't find good figures to estimate the cost of these, I've seen the cost of entire street reconstruction of around 5 000$ per linear meter, but construction may actually be cheaper than reconstruction in this case.

So, anyway, to be able to build affordable housing, you also have to find affordable land, which means undesirable land that is in high supply. Usually, such land is found mostly on the edge of built areas, leading to the "drive 'til you qualify" phenomenon. Alternatively, you have to find a way to cut down on the land consumption per unit, have smaller lots or more units on a single lot.

Since the value of land is linked to the supply of it, it would stand to reason that greenbelt regulations or agricultural land protection laws that forbid development on land restrict the supply and thus make land on the fringe more expensive. There is also a matter of speculation, if land appreciates and taxes on land are low, speculators may believe it best to put off putting their land on the market to let their value climb more. If there are lots of speculators, this can artificially constrain the supply of land too, so putting higher taxes on land to force speculators to put their property on the market faster can help land prices go down.

Building affordable housing... a losing approach?

All in all, building housing for the middle-class ought to be possible, if you select an affordable housing type in a low-value location and/or on a small lot, or low-rise apartments in a likewise affordable location. However, it is hard to find a way to do so for the poor, especially once you take into account transport costs in areas that require cars to get around. In general, it would be best to build along existing rapid transit lines, but these are unfortunately too rare in North America, especially lines that extend beyond the urban core. To really find affordable housing without subsidies, we have to find other alternatives than building housing from scratch.

Still, why is affordable housing for the middle-class, so common in the post-WWII era, so rare nowadays?

The case of the post-WWII entry-level house

The decades following WWII in most of North America, and to a lesser extent, Europe, saw the rapid propagation of affordable single-family houses that would become the homes of an entire generation. Even families with just one income-earner could afford single-family homes. What was the recipe for that era? How did they achieve this?

Well, construction cost-wise, the houses built in that era were not particularly expensive to build, they were cookie-cutter wooden-frame small houses which were often only 800 to 1000 square-foot big, generally sold with unfinished basements or attics to cut down on costs while retaining the possibility for expansions.

Levittown, NY houses, the base models were about 10-meter wide and 8-meter deep, for a floor area of 80 square meters, around 880 square feet, without basement (built on concrete slabs) but with an unfinished attic for future expansion. Many houses have been so modified they're barely recognizable nowadays.
These houses were made as affordable as possible, with the idea that once the main house is paid for, the owners could modify it to fit the needs of their growing family.

Houses such as these are still affordable to build today... it's just a matter of finding cheap enough land.

And that is the single most crucial factor of the post-WWII starter home, which is hard to reproduce. Houses built in this era were built in perhaps the single greatest period of land glut in the history of urban development. This oversupply of land for urban development was the result of technological upheaval in the form of mass motorization and of the rapid creation of the interstate/freeway system. Though the freeways were made to help rapid movement from city to city, they were rapidly taken over by commuters who used them to be able to access cheaper land while maintaining access to the jobs and services of the cities, which then still held most of the jobs and businesses.

Before cars, the fastest mode of travel inside cities was transit, which meant that people had to find land within reasonable walking distance of a streetcar line. That limited the amount of land each streetcar line made available to development and encouraged a still somewhat dense development form...
Streetcar suburb of Chicago
... once the middle-class started acquiring cars, their travel speed increased tremendously. Once limited to 10 mph on transit lines, they could now drive at 20 mph on city streets and even faster on rural roads. Towns that used to be self-contained cities became suburbs of bigger cities as cars reduced travel time to mere minutes between towns separated by 5 or 6 miles. People no longer needed to live in the same city they worked, if land was cheaper in the next town over, they could go there.

Land within easy reach of a major city in the streetcar era, with the streetcar lines in black
In the car era, the land within easy reach of the city has been increased significantly, creating a huge glut of land to develop in the suburbs
The construction of the freeways made this even truer. Now, distances that might have taken a day of walking to cross could be done in 30 minutes. This addition of land to develop on the market brought the price of land way down, at the same time that the Green Revolution made farm lands no longer as essential as before, since the productivity of each acre had been multiplied, and freight could easily feed cities from food grown hundreds of kilometers away.

Though there was a huge movement to metropolitan areas in that era, the amount of land was so great that land was dirt-cheap.

So, small cheap houses with unfinished basements and attics, land made nearly worthless due to the glut brought about by cars and freeways, these are the ingredients for the era of the cheap "starter" home of the post-WWII period.

Where did the starter home go?

The era ended simply because the land glut of the post-WWII era turned into a shortage in many metropolitan areas. While land was cheap, people wasted it with big plots and endless parking lots, but at one point, even the large amount of land started running out. Not only that, but congestion started occurring rapidly because of the spatial inefficiency of private vehicles, reducing travel speed and reducing the land supply. Maintaining speed required building larger and larger freeways and highways, increasing public spending on infrastructure significantly without adequate return on investment. Some areas also implemented greenbelt policies, further constraining land supply.

The dream of a cheap house also came at exponentially higher transport costs, making people depend more and more on oil imported from unsavory overseas regimes. What one didn't spend on housing, one started spending on cars.

It's important to point out that starter homes are still around in second-tier cities that have not maxed their land or which still support the ever-growing freeway system to keep adding land to develop.

Starter home built in 2014 in the suburbs of Indianapolis, still affordable for a one-income family
But what about the existing starter homes built after WWII, aren't they still affordable?

Well, no. The big problem of a starter home is that it's only cheap when it is first built, because it is barebone. Once the owners start making the house theirs to accommodate the growing needs of the family, the house gets bigger and more luxurious. Every addition to the house results in higher market value because of increased desirability. So once the original owners are ready to move out, the house is no longer a starter home, but a big, well-furnished home from which the owner will expect to recover the costs of remodeling.

That is the issue of the "starter home" or "grow home" idea. That home is only affordable once, for its first owners. So for every generation to get its "starter home", every generation has to build entirely new neighborhoods in greenfield areas, where land is cheap. When a metropolitan area matures, this ideal no longer works, the greenfield areas are just too far and are disconnected from the city. So, what can be done?

This is long enough...

I initially wanted to talk about the alternatives for affordable housing in this article, but it's clear it's best if I stop here and write another article. This article only looked at one way of obtaining affordable housing, namely the one way that people tend to obsess about: building housing that is affordable from day one, because of the current supposition that neighborhoods are essentially pieces of art to preserve over the years rather than living and evolving human ecosystems.

For those wanting to read the follow-up to this article, click here.

13 comments:

  1. I'm a cost estimator in Texas for a local production (and occasionally custom) home building company. Additionally, I've worked as a construction superintendent on affordable housing projects, planned communities, and a few customs. Ask me anything.

    ReplyDelete
    Replies
    1. The one thing I'd ask is... have I got everything right? Or close enough?

      Delete
    2. I would say your article accurately describes some of the costs associated with building a home (affordable or not). The “drive till you qualify” rule of thumb definitely holds up today.
      However, there are details that may be worth including in the article, namely those associated with land cost.

      There are a few unseen costs in building further and further inward to a large city. A general rule in the business is “the larger the city the greater the fees”. A bigger city tends to impose larger and more numerous fees. We pay for the right to cut down our own trees, to build our own roads, and to connect to the water supply (they don’t do it for us; they make us pay to do it ourselves). They charge land development fees essentially charging us to do anything to land we already bought and paid for.

      It creates this bizarrely uniform, high base cost with the construction of any house. So, in order to offset this and remain profitable we build larger and less “affordable” homes. That or we build in less demanding, smaller satellite towns.

      It’s a losing game for us too. You don’t make much money at all in affordable housing. We barely broke even. The process of erecting an affordable home can be awfully strange too. People who could live there would have to qualify according to some city criteria, which some would cease to meet midway through construction forcing us to stop building and leaving us in a development limbo. We would have to get city go ahead to enter different stages of construction which would also leave us twiddling are thumbs. Every day you aren’t building you’re losing money. Someone has to be out there, watching, directing, guarding materials, etc. You have to build enough houses to justify the employ of each superintendent. If they aren’t building then they’re a drain on the company.

      I guess that’s all I have to say about that. Hope it was useful. Sorry if I got off topic.

      Delete
    3. Nothing wrong with your article though. Everything you put down was more or less correct.

      Delete
  2. hi Simval84,
    Thanks for demonstrating that Affordable Housing and lifestyles are intertwined. It seems there are two ways to achieve affordable housing:

    1.) adaptable physical housing
    Where prefabricated housing can be purchased cost effectively and upgraded over time. Such schemes would also allow for down-sizing: returning portions (structural boxes) of one's home to be refurbished and shipped elsewhere.

    or
    2.) adaptable locations/dwelling units
    Where one moves around a single neighborhood: starting in a studio apartment moving through a large home, and then returning to smaller empty-nester housing in the same neighborhood.
    and
    Just moving around to be close to where one works: as our jobs change, we may need to move to a different neighborhood to attain the ideal walk-to-work lifestyle.

    The implication is that housing sizes and formats need to be mixed.

    Are there any negative implications to municipalities mandating mixed sizes and formats?

    Thanks

    Joe Lambke
    http://www.animatearchitecture.com

    ReplyDelete
  3. I know this post focuses on greenfield construction, but a thought:

    The stock row house seems to meet all of the criteria of affordability:

    1. It's wood-framed,
    2. It has no complex machinery,*
    3. Its box shape allows for endless reconfiguration based on need.

    In addition, the multi-story row house also has the benefit of serving up to seven 4-person families, which means multiple families bear the cost of land and utility infrastructure. So potentially infill development using the row house form could increase housing supply without increasing housing cost, no?

    *Aside from heat and hot water. The NYCHA used steam heat in its public housing as late as 1970, where the only expensive moving part was the steam boiler in the basement, which required an annual inspection and replacement every thirty years if properly maintained. The iron steam pipes have zero salvage value, and an indefinite lifespan, which also made them attractive given the concerns about tenants stealing expensive copper hot water pipes and flooding their apartments.

    For example,

    ReplyDelete
    Replies
    1. The construction cost is only ever relevant for new constructions. Once housing is built, its value is fixed by supply and demand. It's just that if you can increase supply (build a house) for, say, 300 000$, then similar houses in the area shouldn't be more expensive than that, because then people would just build new houses instead.

      Dense low-rise housing forms are indeed the most affordable type of housing to build. The issue of infill development however is the cost of land. Land is rare in built-out areas, and therefore, expensive. Still, I thin that in general replacing very low-density development by higher density low-rise developments is probably the most affordable way of doing infill development.

      I plan on coming back to this sooner rather than later.

      Delete
  4. I am actually pretty confused about the claim that "All of these are low-rise, walk-up buildings, all offer similar construction costs per square foot to the house models" because... the house models shown have construction costs that clearly go beyond the $100 per square foot value of the later multiplexes. The first one has 161, the second one has 131, and even the absolute cheapest house rises to 124 $ per square foot.

    So what's the deal? Do you really count these as "similar construction costs" as the multiplexes? Premiums of +25% to +60% for single-family homes don't feel negligible to me!

    ReplyDelete
    Replies
    1. Looks to me like he's including the unfinished basements in the square footage totals, on the assumption that it would be relatively cheap to finish later on. Some quick Googling indicates that it might run $25-50 per s.f. to finish a basement. If we assume $40 per s.f., the total cost of the first house with finished basement comes to about $250,000, or just about $100 per square foot total.

      Delete
    2. Indeed, Jacob. Most of the single-family houses have unfinished basements, which can be finished at a cost of 50 $ per square foot, whereas the multiplexes already have finished basements. Once you take that into account, the construction cost of all these structures is very similar per square foot. Which shouldn't surprise anyone, as the structures are built the same way with the same materials, the only thing that changes is how the interior space is subdivided.

      Delete
    3. I see.

      Somehow I expected economies of scale to happen with the multiplexes, compared to the single-family houses. Like perhaps sharing interior walls rather than exterior ones saves on total construction costs, or sharing several water heaters in a single room, or saving on shared electricity and water connections from the building to the street. But apparently none of that matters, as far as construction costs are concerned? That's still somewhat surprising to me.

      I suppose the key thing is that single-family dwellings are systematically larger than multiplex dwellings, such that they don't need as much utility *per square foot*. A single-family house might be twice as large as a multiplex dwelling, but still only use one water connection and one electricity connection anyway, so it balances out as long as we only look at the cost per area rather than at the total cost.

      Delete
  5. Good post. I think I agree. In most (all?) cities with a "housing crisis" the problem seems to come down to a simple matter of over regulation. Zoning codes simply don't permit enough dwellings to be built.

    I am also however concerned about the way large cities seem to pile on costs. I know someone who had paid $40k for a subdivision (just cutting a very generous backyard into another property), before any infrastructure fees (let alone driveway, geo engineering, house, etc).

    $40k for drawing one more line on a map! God knows what they were charged for hooking up sewage, water, electricity, etc.

    For the most part I think it's a simple matter of voter base. Voters in urban regions are nearly always homeowners (tends to be an old, wealthy demographic - the one that owns homes even in highly expensive cities).

    So from the council's point of view the incentive is to shift as much of the running cost (eg renewal of existing infrastructure) onto new residents, away from their voter base.

    IMO the solution is to move up in the chain. The central government (or regional governments) need to get involved in ensuring ongoing/renewal costs are correctly placed.

    But that of course is second place to fixing zoning code.

    ReplyDelete
  6. Can we look at housing affordability without looking at key drivers of homelessness? Financial vulnerability is on a continuum of issues that results in people not having affordable homes.
    Affordable Housing Challenge

    ReplyDelete