Is Missing Middle Really Missing?

There’s en emerging concept within North American urbanism and planning called missing middle. This refers to housing density that’s higher than suburban single-family housing but lower than urban mid- and high-rise buildings. The context is that in some cities with rapid housing construction, especially Toronto, the zoning code is either single-family or high-density, with nothing in between. The idea of allowing more missing middle housing has become a mainstay of New Urbanism as well as most North American YIMBY movements, underpinning demands such as the abolition of single-family zoning in California and Seattle.

Unfortunately, it’s an overrated concept. It applies to Toronto, but not Vancouver or the most expensive American cities, which are replete with missing middle density. The most in-demand neighborhoods have far too many people who want to move in to make do with this density level. Moreover, missing middle density in its New Urbanist form is not even really transit-oriented: low-rise construction spread over a large area is unlikely to lead middle-class workers to take transit when cars are available. The density required to encourage transit ridership and reduce housing costs is much higher, including mid- and high-rise residences.

What’s missing middle density?

A website created by Opticos Design, an architecture firm specializing in this kind of housing, has a helpful graphical definition:

Many of the missing middle housing forms are part of the vernacular architecture of American cities. In New England, this is the triple-decker, a three-story building with an apartment per floor. In Chicago, this is the fourplex, a two-story building with two apartments per floor. In Los Angeles this is the dingbat, with two or three inhabited floors on top of ground floor parking. In Baltimore and Philadelphia (and in London) this is the rowhouse. This history makes it easier to accept such buildings as both part of the local culture and as affordable to the lower middle class.

The triple-deckers in the parts of Providence and Cambridge I am most familiar with have a floor area ratio of about 1-1.5: they have 2.5 to 3 floors (counting sloped roofs as half a floor) and build on one third to one half the lot. A quick look at some Philadelphia rowhouses suggests they, too, have a floor area ratio in that range. Somerville has a population density just short of 7,000 people per km^2, with little non-residential land and some mid-rise and single-family areas canceling out to missing middle density. Kew Gardens Hills has about 12,000 people per km^2, and has a mixture of missing middle and mid-rise housing.

In Continental Europe, the vernacular architecture is instead mid-rise. In Scandinavia and Central Europe the euroblock has 4-7 floors and a floor area ratio of 2.5-4; Urban Kchoze shows many examples with photos, mostly from Prague, and Old Urbanist finds a euroblock in Berlin with a floor area ratio of 4.3. Central Stockholm’s residential buildings are almost entirely euroblocks, and residential density is 17,000/km^2 in Södermalm, 21,000/km^2 in Vasastan, and 28,000/km^2 in Östermalm. Parisian density is even higher – the floor area ratio of the traditional buildings looks like 4-5, with about 30,000-40,000 people per km^2.

Is missing middle really missing?

In Europe the answer is obviously no: lower-density cities like London are largely missing middle in their inner areas, and higher-density ones like Paris have missing middle density in their outer areas. But even in North America, where the term is popular, the expensive cities where people call for abolishing single-family zoning have missing middle housing. In addition to the above-listed vernacular examples, New York has brownstones all over Brooklyn (the term Brownstone Brooklyn refers to the gentrified inner neighborhoods, but this density is also seen in outer neighborhoods like Bay Ridge and Sheepshead Bay).

Vancouver is an especially instructive example. English Canada’s big cities are fast-growing, and a zoning regime that’s historically been friendlier to developers than to local NIMBYs has encouraged high-rise growth. Moreover, the high-rises are built in the modern boxy style (earning the ire of people who hate modern architecture) and tend to target middle-class and high-skill immigrant buyers (earning the ire of people who blame high housing costs on new construction). In contrast, vast swaths of Toronto and Vancouver are zoned for single-family housing.

And yet, Vancouver has considerable missing middle housing, too. The population density in Mount Pleasant, Fairview, Kitsilano, and West Point Grey is similar to that of Somerville and Eastern Queens. Buildings there are in modern style, but the housing typologies are not modernist towers in a park, but rather mostly buildings with 2-4 floors with the medium lot coverage typical of missing middle. I lived in an eight-unit, three-story building. Across from me there was a high-rise, but it was atypical; for the most part, that part of Vancouver is low-rise.

Shaughnessy offends people in its extravagance and wealth. In one Twitter conversation, an interlocutor who blamed absent landlords and foreigners (read: Chinese people) for Vancouver’s high housing costs still agreed with me that Shaughnessy, a white Canadian-born single-family area, shares the blame with its low-density zoning and very high residential space per person. Legalizing accessory dwelling units (“granny flats”) and townhouses in such a neighborhood faces local political headwind from the neighbors (who are still nowhere near as empowered to block rezoning as they would south of the border), but not from citywide social movements.

And yet, the density in the inner Westside neighborhoods near Broadway and Fourth Avenue is insufficient, too. It’s of course much higher than in Shaughnessy – I never really missed not owning a car living in Kitsilano – but the price signal screams “build more housing in Kits and Point Grey.”

Is missing middle transit-oriented?

Not really. In Providence the answer is absolutely not: car ownership is expected of every person who can afford it. The nearby supermarket, East Side Market, has an enormous parking lot; I’d walk, but it was obvious to me that my mode choice was not the intended use case. Even some Brown grad students owned cars (though most didn’t); at Columbia, car ownership among people below tenure-track faculty rank approaches zero. Once they own cars, people use them to take trips they wouldn’t otherwise have made, reorienting their travel patterns accordingly.

In Cambridge, car use is lower, but still substantial. The same is true of Vancouver (where outside Downtown and the West End the entire region’s density is at most missing middle, even if the typology is towers in a park and not uniformly low-rise). In Kew Gardens Hills, people seem to mostly drive as well.

This is not a universal feature of the urban middle class. In Stockholm, my postdoc advisor as far as I can tell does not own a car, and commutes to work by bike. Both there and in Basel, biking and using transit are normal and expected even among people who earn tenured academic salaries. At 7,000 people per km^2, people can forgo driving if they really want to, but most people will not do so. Only at the higher mid-rise density will they do so.


  1. Michael James

    Is missing middle transit-oriented?
    Not really. …
    At 7,000 people per km^2, people can forgo driving if they really want to, but most people will not do so. Only at the higher mid-rise density will they do so.

    I think you have addressed the wrong question. Surely the answer should be: Yes, at 7000/km2, it can be efficiently served by rapid transit (ie. not just buses). Also if a whole quartier of say about 5km2 is of that consistent density then it will autocatalytically generate a walkable area, ie. with retail, services (medical etc) and recreational (restaurants, sports facilities etc). However, to generate this you will need housing of the type on the right half of that diagram. Also I don’t think the definitions are adequate. I actually refer the range 10-15 floors as mid-rise and above that as hi-rise. I believe Haussmannian-style (or label it “Euro-core”, 6 to 8 storeys) needs its own descriptor because it really falls into a category of its own with respect to functionality and urbanism, eg. the original Haussmannian calculation of maximum height as a proportion of street width, and still accessible without mechanical elevators (of course today they will have an elevator but to meet OHS does not need more than one), and fewer issues w.r.t. safety (a single set of stairs) and construction (cheaper, quicker, simpler; eg. suitable for CLT).

    The context is that in some cities with rapid housing construction, especially Toronto, the zoning code is either single-family or high-density, with nothing in between.
    Unfortunately, it’s an overrated concept. It applies to Toronto, but not Vancouver or the most expensive American cities, which are replete with missing middle density.

    The problem with that is that the definition of “missing middle” is far too wide, ie. the left-hand half of that diagram is not especially consistent with achieving overall densities adequate to achieve walkability, transit viabiity etc. And in turn, the problem with North America and most of the Anglosphere is the way it is a dog’s breakfast of all these typologies. In Australian middle-to-outer (not exurban) suburbs you’ll find all those types side-by-side and it’s an ugly mess; post-war SFH might be demolished to be replaced with what we call six-packs, while today an old bigger suburban block is redeveloped with townhouses which are more aesthetically acceptable (and sold to owner-occupiers at higher prices).

    Being driven by property developers it does have a tendency to hi-rise (or what I call mid-rise, ie. >15 storeys) while those lower structures have often been built by small builder-developers who purchase stock plans off the shelf rather than architect-drawn, project-specific ones. This is the logic and origins of the dramatic contrast between single-family and mid- and hi-rise. In fact a common ploy by developers is to submit plans for mid-rise then once given building approval immediately submit a modification to turn it into much higher. I contend that this lack of proper planning control–its corruption by developers–lies behind a lot of NIMBYism. Where I live, an inner-fringe area that had a legacy of light-industrial, but quite attractive, brick buildings of 4-6 floors converted into “loft” apartments, there were years of public consultation etc arriving at a very good redevo plan for a big brownfield site to match this height limit. The first two, very upmarket riverside apartment blocks, adhered to this plan and are really good at 6 & 8 floors, but then for every subsequent building the developers have claimed the economics had changed and they demanded higher buildings on the exact same sites so they have reached 25 floors and the remaining riverside site is on the books at 45 floors! A cynic may well think this was a strategy cooked up between the developers and the city council all along. Why should anyone trust this sham of a planning process? I imagine it is the same all over North America.

    • Alon Levy

      Haussmannian density is what I call mid-rise. Brick structures (or sometimes concrete podiums with wood in the US) are common, and elevators are useful but (at the lower end) not absolutely necessary. Here the range is 5-9 floors; in New York it’s 5-10, since there are a lot of New Law tenements built up to 10 that look like the buildings here would with one extra story and no gabled roof. In Central Europe the euroblocks are somewhat less dense – the lot coverage is 50-60% whereas here it’s more like 60-80% – but the height is similar, and the density is not much lower than what you could see here.

      The missing middle density I’m describing is walkable, but only up to a point. Providence is walkable in the sense that the streets are pleasant to walk on when there’s no snow, but there’s not enough density to be able to comfortably walk to stores. Walking to school at such density is possible but questionable, depending on class size (and American schools tend to be big, esp. in high school, esp. in cities). London is denser, but walking distances to the Tube are still pretty bad. TfL says that London has 320 cars per 1,000 people (New York: 230).

      • Michael James

        I understand but my point is that ‘mid-rise’ to some will mean some significantly taller buildings, up to 15 to 20. My concern is to define a building type that has a particular functional meaning–all those things associated with Haussmannian buildings. Note that strictly one should define these types by actual heights rather than storeys–thus you’d find that those NYC buildings of 10 floors could well be within Haussmannian limits since everywhere today we’ve reduced “modern” ceiling heights to the absolute minimum that some expert, or more likely developer, has deemed is all we need. Thus a high-ceilinged Haussmann building of 7 floors could be 10 floors of a modern building. Actually, to be strictly correct, “Haussmannian” is probably not quite correct since the last major change to building code in Paris was in 1902 when, on wide streets, using a different profile of the roofline above the 6th floor; this allowed up to three mansard (attic) floors instead of the previous single floor. Most of the Haussmannian buildings on the major boulevardes had extra floors added to them after 1902–including most of rue de Rivoli and Avenue de la Opera etc. (I’m looking at Norma Evenson’s book which uses these streets as illustration, plus the Hotel Lutetia which has 9 floors.).

        Anyway, the point is that this limit (of 28-30m but with setbacks above ≈20m) was arrived at over centuries of evolution in Euro cities, particularly Paris. I lived on Ile St Louis which was mostly developed over a short period in the mid-17th century–it was all over by about 1660–predating Haussmann by 200 years! Putting my geneticist hat on, I contend that this is part of why the form is so good, and why Paris is commonly considered the most beautiful city and by all kinds of people who don’t consciously understand any of this, because it embodies some basic human appreciation/quality of scale and proportion etc. And the real point of course is that this form achieves the highest density–especially over a wide area (≈100km2)–and ipso facto there is no need or point in building higher!

        The missing middle density I’m describing is walkable, but only up to a point.

        Right. I’d say that 7,000/km2 is at the lower limit of creating walkable zones. But again my point is that they become walkable because the density automatically creates it (as long as zoning code etc doesn’t stop it). Like you can step outside your apartment in the 12th arrondissement and be only a few steps from a boulangerie, cafes and hundreds of all kinds of urban facilities, and also be close to a food market and a medium size supermarket, and schools too. That is, while such a zone may have a diameter of say, 1-2km around a transit hub, that doesn’t mean you will have to walk anything like that for most things as the density will support all those services everywhere.

  2. IAN Mitchell

    Alon, I think you’re missing the point of missing middle as an identifier.

    Missing middle is present in older cities BECAUSE THEY ARE OLDER.

    If you take for example, my new home, and America’s most rapidly urbanizing city, Austin, there’s basically nothing between single-family homes and 5+ story condos, particularly in new construction.

    The same applies in Orlando, Tampa, Jacksonville, Miami, Fort Lauderdale, Dallas, and other sunbelt cities.

    Missing middle is most illegal or absolutely unaffordable (due to financing and opaque regulations) to BUILD, Texas Doughnut is pretty easy. Tract homes are really easy.

    A modern equivalent of a polish flat? Won’t see it.

    It may not be “missing” in many places in terms of built environment, but it’s absent from new construction.

    As far as Providence, I don’t blame a lack of residential density. I blame a lack of infrastructure. You have commuter rail and mixed-traffic buses. The housing isn’t missing a middle, but transit is.

    Of course, the fiscal conservative in me thinks that Providence is better suited to building protected bicycle infrastructure than pouring any real concrete in the near-term.

    What also bears examination is whether Providence is suffering from Cincinnati syndrome: is everything old illegal? Do the new building codes attempt to enforce suburban standards on a prewar city?

    Are grocery stores providing parking to attract customers, or because they’re being compelled by force of government to provide it?

    • Alon Levy

      My main comparison for Providence is Tel Aviv. The public transit in Tel Aviv is shit; it’s the largest first-world non-US city that doesn’t have rapid transit (a subway or an S-Bahn), about 50% larger than the second largest, which is Leeds. The overall density in Tel Aviv proper is about 8,000/km^2, but that includes low-density suburban neighborhoods and commercial areas. The neighborhood I grew up in, the Old North (District 3), has around 16,000 people per km^2, and five figures of which the first is a 1 are common in some residential inner suburbs like Ramat Gan. With that density, people routinely walk to retail, unless they’re going out to a big shopping mall. Somerville’s density seems borderline for this behavior; the East Side of Providence is just not dense enough.

      I don’t know the building codes in Providence, but I don’t think they matter that much. In a city whose tallest building is empty, there isn’t going to be new construction. I bring it up as an example of uniform missing middle density and not as an example of recent missing middle development. You’re right that if you only look at recent construction the missing middle really is missing (though the dingbats were built in the 1960s and 70s – it’s not just a prewar thing), but that’s specifically because the metro areas with the strongest demand today have demand for higher density or else space for single-family sprawl.

  3. Pingback: Today’s Headlines – Streetsblog Los Angeles
  4. Benjamin Recchie

    ” In Chicago, this is the fourplex, a two-story building with two apartments per floor. ”

    Say what now? I’ve lived in Chicago almost all my adult life, and not only have I never heard of a fourplex, I have trouble imagining where one would be in the city. The dominant middle-density buildings here are two-flats and three-flats (two or three story buildings with one apartment per floor, and perhaps a basement apartment). See

  5. Connor Harris

    Lots of interesting points here. A few disconnected comments (apologies if this is a double-post; the comments were a bit glitchy for me earlier):

    First, I have to agree with Ian Mitchell above that there is a missing middle (perhaps defined more restrictively than in the illustration) in recent construction, if not in the extant housing stock. To take Ian’s example, from 2000 to the 2012-16 five-year American Community Survey window, the number of units in 2- to 4-unit structures in Austin increased by 8.3% (thanks to zoning and to pointless “compatibility requirements” that apply to developments of three or more units), while population increased ~40%. Multi-unit construction smaller than 20 units on the San Francisco Peninsula is essentially nonexistent. You can check out Issi Romem’s maps of housing construction for some nifty illustrations. (The Census Bureau tabulates housing units by size of structure; a search for “units in structure” in American FactFinder should bring up the relevant tables.)

    Second, I think that in areas like Silicon Valley where employment is relatively dispersed, more lowish-density housing could do a lot of good. Atherton should be turned into central Hong Kong on general principle, but aside from that, the contiguous urbanized area of Santa Clara County from Palo Alto to south San Jose is about 600 km^2. If even two-thirds of that area is single-family residential at a typical density of 6 units/acre ≈ 1500/km^2), then increasing a quarter of the lots from 1 to 2 units would meet the county RHNA requirements twice over. Transportation would, admittedly, be a problem, though even marginal upzonings of the areas around tech hubs could let a nontrivial number of workers live within bicycle range, and might make additional park-and-ride bus service viable.

    Third, one argument I’ve seen for low- and middle-density housing is that it’s cheaper to build and maintain than higher-density housing, with a big jump in prices at the transition from wood to metal construction (usually about 4 stories, I think, though I could be wrong). This forms a centerpiece of Randal O’Toole’s argument that SFR-only zoning in established areas is completely irrelevant to housing affordability (not an exaggeration, sadly; see page 21 here). Obviously O’Toole et sim. take this point to absurd lengths, but construction costs are, perhaps, a minor consideration for markets less insane than California or Boston.

    Finally, in May, I attended an event in San Jose on missing-middle housing. Some interesting claims from the discussion, not necessarily endorsed:

    – Even under liberalized zoning, new Bay Area housing, at least at first, would be so upmarket that essentially all of its inhabitants would have cars.
    – The ADU bill made it through the California legislature with support from homeowners who could imagine wanting to build an ADU (to house relatives, for example). Stand-alone housing developments don’t have this constituency.
    – It is possible to sue a housing project under CEQA on the grounds that it will make parking harder.
    – Construction unions will use bogus CEQA lawsuits against infill housing developments to extract union-labor guarantees. Out of self-interest, they don’t do this to ADUs, as most ADU construction uses some union labor, and a lot of homeowners couldn’t afford to build an ADU with union labor exclusively.

    • Alon Levy

      Two big things:

      1. In California, the limit of woodframe on a concrete podium is 7 floors. The big cliff in construction costs is from that to taller buildings. This is why SB 827 picked its heights: the highest end is exactly the state’s limit for woodframe based on earthquake design standards. It’s firmly in mid-rise territory and not missing middle. I wouldn’t even call Tel Aviv’s 4-story buildings, which are FAR 2, missing middle. The density they provide is around 15,000 people per km^2 at what is locally considered middle-class comfort (120 m^2 for a family). Elsewhere, you don’t really see big cliffs; in Canada there is a gradual increase in construction costs with height, but it’s not steep.

      2. “Transportation would, admittedly, be a big problem” is a huge understatement. There is one rapid transit right of way through the Peninsula, which doesn’t go near any of the big Silicon Valley employers. Missing middle density spread all over the area would not have the town center-periphery structure common in various European suburbs that get away with low density and high transit ridership by concentrating higher densities in a block or two next to the station. Nor is there a strong grid of arterials for buses; the only arterial worth running buses on, El Camino Real, is parallel to the Caltrain corridor. People would either drive or ride TNCs. Biking is semi-plausible for short distances, but the land use in Copenhagen and Amsterdam is mid-rise and not missing middle, so I don’t know of any examples of mass bike usage at missing middle density.

      Smaller things:

      3. “New Bay Area housing would be so upmarket that essentially all of its inhabitants would have cars” seems false given recent increases in transit mode share in the Bay Area. I’ve seen people plausibly explain transit usage decline in Los Angeles as a byproduct of gentrification displacing poor people (i.e. essentially the only transit users in LA) from dense neighborhoods, but BART ridership kept rising until 2016 or 2017 and the region’s transit mode share is still rising as of ACS data with a few years’ lag. Caltrain ridership I think is still increasing, not to mention the tech shuttles. Tech workers in general have very low car usage relative to their incomes, on a par with academics or maybe even less.

      4. People here and on Twitter keep telling me that missing middle is not for California (where mid-rise probably pencils literally everywhere in SF proper and anywhere in LA near a train station) but for Austin and other places where a union member is considered a type of terrorist.

      • Michael James

        Alon Levy, 2018/09/11 – 13:26

        1. In California, the limit of woodframe on a concrete podium is 7 floors. The big cliff in construction costs is from that to taller buildings. This is why SB 827 picked its heights: the highest end is exactly the state’s limit for woodframe based on earthquake design standards. It’s firmly in mid-rise territory and not missing middle.

        Excuse me banging on about it, but this exactly points out what I called attention to earlier. The use of the term “mid-rise” for 7 storey buildings is seriously misleading and counterproductive, especially in an era where the industry is building supertalls at 80+ storeys. Under this/your definition (or your acceptance of this definition) Paris was building mid-rise in the 17th century.
        Also, I am pretty sure that SB 827 specified “four- and five-story multifamily housing within a half-mile of transit “. In comments I noted that the Euro-block (or Haussmannian) limit of 7 storeys was more appropriate, and all the more so if that is the CA limit for woodframe. That is, these two things nicely converge: building code (even though with modern tech the real limit for CLT timber buildings is higher) and the proven urbanism of this scale, plus the fact that it easily surpasses any density that any American area is likely to achieve.

        Otherwise, your response to Connor Harris is perfectly correct. Doubling existing suburban housing density everywhere is an absolutely atrocious concept, not least because it greatly exacerbate road congestion (it cannot possibly bring enough people to within biking/waking distance of their workplaces) and it wouldn’t solve the unaffordability issue. It wouldn’t create any kind of good urbanism. By contrast a TOD zone of 1.6 km radius (8km2) would hold the following number of residents:

        density (/km2)….total residents
        25k (Paris)…………201,000

        This would represent a small slice of most Bay Area counties, and would create its own urbanism (and mini-CBD) around those BART – Caltrain stations. Absolutely zero requirement for hi-rise or what I call mid-rise (>10<20 floors). The 4-5 floors of SB 827 would struggle to achieve the 10k, and as you point out it would struggle to support true urbanism (walkable fully-serviced zones), and it would not be as economic as the 7 floors which could achieve 15-20,000/km2.
        Re the discussions about car use: I point out again that car ownership is not the same as car usage. In Paris I am struggling to think of anyone I knew who didn't own a car (including myself for the first year when I decided it was ridiculous). But equally I didn't know anyone who lived in intramuros-Paris who used their car very much and especially used it within Paris. For obvious reasons, one of which is not road congestion (surprisingly but one seems to be able to zip around pretty niftily most of the time) but more the parking problem and the fact that almost everyone keeps their car garaged quite separate, and distant, to their residence. Plus the huge factor that you can get around so easily and quickly by other means: Metro, buses, walking, velib. And no one drives their kids to school, mostly because it is within easy walk but even if across town then the kids start using the Metro or buses quite young.

        • Untangled

          There’s still a difference between a regular high-rise and a skyscraper (150m+). You don’t need it to be supertall or 80 stories+ for something to be considered high-rise. I think 7 stories is an okay definition of mid-rise, anything that’s within a comfortable walk-up would be low rise and anything higher than that is mid-rise until you get to around 50-ish meters.

          • Michael James

            Untangled, 2018/09/12 – 00:39

            I think 7 stories is an okay definition of mid-rise, anything that’s within a comfortable walk-up would be low rise and anything higher than that is mid-rise until you get to around 50-ish meters.

            Then you are actually agreeing with me. The vast majority of Euro-block (Haussmannian) buildings were walk-up (in the two apartments that I spent most of my time in Paris, both were 5th floor walk-ups) and so you are agreeing with me that they are more “low-rise” than “mid-rise”. The limit is about 30m which can be up to 10 (low ceilinged) floors. That is my threshold for the beginning of “mid-rise”. They are essentially the same as 4 to 5 storey buildings, just a bit taller, which is why labelling them ‘mid-rise” is misleading. Building codes mean that anything higher will be mandated to have two elevators and two independent fire-stairs; in turn this will tend to force bigger bulk for both capital cost and maintenance cost (elevators are a major element of running costs in apartments).

        • adirondacker12800

          Wood burns. The Japanese were building 15 story buildings outta wood centuries ago. It burns. Effectively in many jurisdictions the fire code means nothing higher than 4 stories and higher than that, no wood. It burns. Other things like wiring regulations get much tighter too. Did I mention that wood burns?

          • Michael James

            adirondacker12800, 2018/09/12 – 02:16

            Wood burns. The Japanese were building 15 story buildings outta wood centuries ago. It burns.

            That is seriously out of date. The CLT and other engineered timbers are fire-resistant, and actual large solid-timber structural elements such as verticals or girders take a long time to burn, as their outer surface charcoals and protects/slows down its own destruction. Once completed a modern timber structure is relatively resistant to fire and more importantly, provides more than mandatory time periods for evacuation of the building. It is most susceptible while under construction as in the recent fire at a modern timber 5 storey building that got a lot of publicity.

            In case you have only just woken from a 20-year sleep (jiving off your moniker …) the most dramatic fires in the past couple of years have been hi-rise concrete, steel and glass buildings. Grenfell Tower with 70+ lives lost.
            The part of a fire that kills you is the stuff inside the building that creates smoke and toxic fumes that asphyxiates you before you can escape. Even the houses lost in the bush fires in California, Australia and Europe–do you know it matters not a jot whether the house is timber or masonry, because a house burns from the inside out, almost always by embers gaining access by breaking glass windows. You can actually protect a house by nailing timber boards over the windows etc.

            Those Japanese “buildings” are not real buildings but ceremonial towers.
            In contrast those 6 storey apartments on Ile St Louis have housed almost 100,000 residents for over 360 years, continuously. I suppose there must have been some fires but I am not aware of any. FYI, the creation of Ile St Louis out of the Ile aux Vaches etc by Henri IV and his chosen engineer-builder Christophe Marie, was the first systematic attempt to create a coherent housing format for Paris, to build both functional, long-lived and beautiful residential buildings. Prior to this most of Paris was a part shanty town of poor-quality all-timber buildings in poor repair, or too-expensive all-stone palaces. (This new style is only faced with pierre-de-taille.) It was very successful and became the model for Paris’ future; 200 years before Haussmann, which you can read about in Jean Dejean’s How Paris Became Paris: The Invention of the Modern City
            En plus, these buildings have a lot of plaster (old fashioned, not thin plasterboard), mortar and stucco; plus the inter-floor space is filled with something similar (some kind of stucco) which at >20cm deep provides both noise insulation and fire-resistance.

          • Eric

            Modern fire codes require sprinklers and other measures which are effective in preventing fire.

            Interestingly, a lot of these modern wood buildings do burn down – when they are under construction so the fire prevention systems are not yet in place. But not when they are finished and occupied.


          • adirondacker12800

            Engineered lumber ain’t cheap. In this part of the world commercial construction has had to be “fireproof”, it’s not, for decades and decades and steel studs and 5/8th inch drywall with armored cable/conduit wiring has been used for decades, the contractors know how to do it cheap. And that’s how they build multifamily, cheap. Yes. the tall Japanese buildings were ceremonial towers because nobody wants an apartment or office on the 12th floor when there are no elevators.

          • Michael James

            adirondacker12800, 2018/09/12 – 11:42

            Engineered lumber ain’t cheap. …. cheap … cheap … cheap…

            Nevertheless, for those constructions up to 10 floors that have actually been built, they are 30% cheaper than standard concrete and steel (and plasterboard).

          • adirondacker12800

            With same fire rating on the walls and doors and emergency egress and the same wiring specs and… and the people who make the fire blocking materials certified that works with engineered lumber and …. and… I can make buildings reall cheap. People die in them.

        • Alon Levy

          SB 827 had a few different limits. The lowest was 40 feet, the highest was 85, depending on distance from transit and on whether it was a bus or a train station. I vaguely recall Brian Hanlon telling me that in California’s codes, 85′ is the maximum beyond which all construction must be fireproof and made of concrete. The rest of the US has similar rules, although less stringent because no earthquakes. The result is that American developers almost never build anything between 8 and 14 floors; the 70′ or 85′ limit is such a big break in construction costs that if you go over you might as well build 15+ floors.

          This side of the Pond, it’s useful to look at the maximum normal height in Barcelona, prewar Paris, etc., as its own break point. You do get a lot of 10-to-14-story developments here, but they’re HLMs and they don’t build any of that anymore as far as I can tell. It’s about history and not building codes, since this part of Europe ran out of wood in the High Middle Ages so everything is made of brick or concrete, but it fits.

          • Michael James

            Alon Levy, 2018/09/12 – 04:18

            The result is that American developers almost never build anything between 8 and 14 floors; the 70′ or 85′ limit is such a big break in construction costs that if you go over you might as well build 15+ floors.

            Right. As it happens (but no accident) that is the same as Haussmannian limits. And it is consistent with my claim, ie. that this height is a sweetspot. It doesn’t require as deep or expensive foundations. It is much quicker to build (especially as most is prefabricated) and these features add up to a lot of savings (a big part of the cost of hi-rise is the cost of financing for their long builds). That jump from 8 floors to 15 and 25 floors in the brownfield site redevelopment near me, is what you expect.

            Still, those building codes you speak of are most certainly behind the times. Tall timber towers are being built or planned everywhere. As I think I have written on your blog, one of the first, the Forté building in Victoria Harbour, Melbourne Docklands, is interestingly of Haussmannian dimensions: 10 storeys and 32m high. Also not too far from me is another 10-floor timber building, “25 King Street”.

            Re loss of timber and forests in Europe, that is only partly true. It’s true that there is limited local hardwood but I’m not sure if they don’t use fast-growing timbers in CLT. A lot of the world’s CLT for these (hi-rise) projects currently comes from Austria. The switch away from timber as building material was mostly driven by the enthusiasm for concrete and steel.

            The city of Bordeaux in southwestern France has pledged to build 270,000 square feet of wooden spaces per year for the next 15 years. One of the projects leading the way is the Hypérion tower, an 18-story residential building with CLT floors and walls. It aims to be one of the tallest timber structures in the world when completed in 2020.
            The Hypérion project has an additional goal: to promote the use of locally sourced wood from the Nouvelle-Aquitaine region that surrounds Bordeaux. The wood industry has lost ground in France, accounting for less than 5 percent of construction, compared to as much as 25 percent in the Nordic countries. The Hypérion tower will be a showcase for chestnut and oak from the Périgord, to the north of Bordeaux, which make up building’s facade and beams, while the panels that secure the facades are made from pine from the nearby Landes forest.

            The Landes forest of 10,000km2 is supposed to be the largest continuous forest in the EU (not sure if this claim predates some of those eastern European countries joining–some of them have big forests). It was drained from swamp beginning in the 18th century and planted with pine to supply the timber and paper industries.

            I am not entirely convinced by the enthusiasm for taller and taller timber buildings. But I think it makes a lot of sense for Haussmannian scale buildings.

          • Michael James

            Oops, in my earlier post I see I made a blunder. Ile St Louis doesn’t have 100,000 residents! It is approx. 100,000 m2, ie. 10 Hectare; with 4,453 residents (last census) to give density of 44,530/km2 (one of the highest in Paris and in fact the world; except for the church it is all residential).

      • Matthew

        Just about every city and town in the Netherlands has strong cycling mode share, especially compared to anywhere else in the world. I’m sure you can find lots of examples of missing middle out there. Don’t just look at Amsterdam. How about Houten, which is a suburb new town of Utrecht? There’s a good study from ITDP about the place.

  6. Alai

    San Francisco has a lot of the so-called missing middle as well. And yes, it’s very expensive.

    But I think your analysis is backward. The fact that the few older cities that have it are so expensive and desirable just goes to show that people want to live in those neighborhoods. So expanding the amount of missing middle would take some of that pressure off while providing a lot of new housing at a reasonable construction cost.

    I think it’s plenty transit-oriented too. The problem is not the density, it’s that the government policies have heavily favored cars to the point that it causes all sorts of problems, but the favoritism remains, and drives people to own cars. Eg parking minimums for new construction, large amounts of public space being used for free parking, and not charging car owners for externalities (some progress has been made on the first one in SF, but not so much on the others). Tragedy of the commons and all.

    • Alon Levy

      The “it’s desirable, therefore we should build more of it” line runs into a fundamental problem: the main determinant of housing prices is not housing typology. Are single-family favored quarter suburbs expensive because secretly everyone wants to live in SFR-zoned areas? No, they’re expensive because people want to live in the favored quarter and zoning laws make it hard to build more there. By the same token, you can find expensive missing middle neighborhoods (Cambridge and Somerville on the Red Line, Kits and Point Grey, Hammersmith), expensive mid-rise neighborhoods (the Village, the 16th Arrondissement), and expensive high-rise neighborhoods (the Upper West Side, Bukit Timah).

      • Nathanael

        But in the US, neighborhoods with “missing middle” housing are *uniformly* expensive and have near-zero vacancy rates. In any city doing better than Detroit, *all* the rowhouse / duplex neighborhoods are expensive. There are serious signs of a shortage of the type. Whereas there are lots and lots of “single family only” zones which are unpopular and disfavored. (And even high-rise zones with high vacancy rates.)

        This is probably simply because it’s illegal in nearly every city in the US to build more of the “missing middle” housing. Legalize it, and a few single-family neighborhoods will convert, and the pricing will normalize — as it has in Vancouver, frankly.

        You seem to be overlooking the point. The missing middle is, in fact, really missing from new construction, and there’s unfulfilled demand for it. And it’s missing because of zoning.

        • Nathanael

          Now, is this type of housing going to encourage public transit use? Maybe not; you may be right that it’s naturally auto-oriented. But it’s certainly in high demand and missing.

  7. adirondacker12800

    This is overly simplistic. Cars happen. You don’t have to live in a trolley suburb anymore they stop building trolley suburbs. It’s much more complex but cheap cars and cheap gas makes all the other more complex reasons possible for masses of people.

  8. liberallandscape

    Even higher-density neighborhoods can be quite automobile-oriented. See my post at
    for one example (and a reference to some academic work on this subject). In many Third-World cities, the highest-density neighborhoods are also the most well-off–and the most car-dependent. There isn’t as simple a relationship between automobile use and density as one might expect.

    Chris Winters

    • Alon Levy

      From your post:

      In 2010, there were 6457 people living on the first six or seven blocks of Wilshire Boulevard east of Glendon Avenue. The population density of these blocks was 16,150 people per square kilometer (41,828 per square mile), which is pretty high.

      6,457 people isn’t a lot. It doesn’t support much retail, it doesn’t support an urban American school, and it doesn’t support any job cluster. In theory it could be transit-oriented if there were a subway on Wilshire, but between Condo Canyon and Century City, Century City is much more important to serve, so the Wilshire subway plan veers from the street it otherwise follows and hits the job center.

      • Chris Winters

        Actually, there will be a subway stop, someday, at Wilshire and Westwood, right in the area about which I’m writing. But I think you miss the point, which is that there just isn’t as simple a relationship between population density and prevalent transit mode as you imply. Many high-density apartment areas in the world are quite automobile-dependent, partly for cultural reasons, partly because the geography of the urban area in which they’re located limits the usefulness of public transit.

        Chris Winters

  9. marvin gruza

    what “can be efficiently served by rapid transit (ie. not just buses)”
    How will new transit technology alter this issue? (and related assumptions, understandings and formulas)
    If self drive/on demand vehicles can shuttle users to a train station that is otherwise either too far or a strain to walk to with a total time of call/request -to station drop off of 5-12 minutes, will a much larger but less dense area be able to be efficiently served? How certain are we that is this future? How do we plan for it now? How do we implement transit projects knowing/or assuming that his is down the tracks?

    Self drive/on demand vehicles will:
    *eliminate space used for parking
    *increase capacity of roads as the space used by needed space for driver reaction is reduced
    *decrease travel time even on non grade separated roads as intersection and all roads will have increased efficiency

    • Michael James

      marvin gruza, 2018/09/11 – 20:43

      How will new transit technology alter this issue?

      No one really knows. But history suggests that once commuters are in cars they have a strong tendency to keep going all the way to their destination, rather than make a mode change. Especially when paying twice for transit. Or three times if you have to use another AV at the other end too. It’s why walkable neighbourhoods are so popular (including with Americans in the rare cases they get to experience them).

  10. payton

    The Census data set that most closely tracks “missing middle housing” is housing unit completions in 2-4 unit buildings. New units in such buildings have fallen OVER 90% since the early 1980s:

    You’ve chosen a few (North) American cities which actually have “missing middle,” while ignoring that the type has vaporized across most of the rest of the continent. (Boston, I believe, has the highest proportion of its residents in 2-4 unit buildings of any large US city.) These types are essentially illegal in the South and West and in suburban areas (where most population growth occurs); they’re also very difficult to build new even in large Northern cities where they were historically common.

  11. Michael

    The Marina District in San Francisco gets a lot of grief for its stereotypical demographic, but it’s a prime example of a mix of densities in the middle ground. If you’re unfamiliar, take a googlearound…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.