"it is the best long term choice"

It's certainly a good idea, but we're not living in the long-term: we need solutions in the next 10-20 years.

"surprisingly affordable"

Not as a single solution: to get the same reduction in oil and CO2 emissions, in the same timeframe, would be much more expensive than a PHEV/EV conversion.

"The statistics you seek for modal share were generally simply not kept..."

Then you have to be somewhat restrained about what you claim for it. If you don't have the data...well, as we'll see in a later post,you can often go very, very wrong.

" Boston built their first subway (using streetcar rolling stock) in 1897, and various other subways later"

A fair point - certainly, subways are part of what you're talking about. OTOH, what I was talking about was tram/trolley, and I believe Boston had finished replacing it's horse-drawn cars by 1900. That tells us something about the other 1,000 cities, most of which didn't have subways.

"you write long, multi-point posts"

Yeah, this kind of thread gets difficult to follow, doesn't it? I may well miss something...

"Contrary to common sense and physics. I forgot what examples you found but they were "unconvincing". I SIMPLY DO NOT ACCEPT such falsity."

My, oh my, oh my.

This is an extremely good example of where strongly felt intuition ("common sense") can lead you badly astray. As I noted above, we need data, and the data in this case is very different.

So, is urban housing more efficient?

No.

Here's the data:
Residential Energy Consumption Survey http://www.eia.doe.gov/emeu/recs/contents.html
2005 Residential Energy Consumption Survey--Detailed Tables http://www.eia.doe.gov/emeu/recs/recs2005/hc2005_tables/c&e/detailed_tables2005c&e.html
Average Consumption, British Thermal Units (Btu) per Household (US9)http://www.eia.doe.gov/emeu/recs/recs2005/hc2005_tables/c&e/pdf/tableus9.pdf
Housing Unit Characteristics and Energy Usage Indicators (US1:Part 1)http://www.eia.doe.gov/emeu/recs/recs2005/hc2005_tables/c&e/pdf/tableus9.pdf

Total BTU's
Type of Housing Unit Single-Family Detached. 108.3
Single-Family Attached 91.7
Apartments in 2-4 Unit Buildings 84.5
Apartments in 5 or More Unit Buildings 53.8

Floorspace/ Household (sqft)
Type of Housing Unit Single-Family Detached. 2,720
Single-Family Attached 1,941
Apartments in 2-4 Unit Buildings 1,090
Apartments in 5 or More Unit Buildings 872

BTU's/ thousand SF
Type of Housing Unit
Single-Family Detached. 39.816
Single-Family Attached 47.244
Apartments in 2-4 Unit Buildings 77.523
Apartments in 5 or More Unit Buildings 61.697

We see that Single-Family Detached are most efficient, followed by Single-Family Attached (townhouses), then Apartments in 5 or More Unit Buildings, and at the bottom are Apartments in 2-4 Unit Buildings!

We see that apartments only manage to use somewhat less energy by being much, much smaller on average.

Why is this? This is likely due primarily to the fact that heat loss and gain are much more affected by windows than by outside wall exposure, and apartments and condo's maximize the outside exposure and window area for all rooms.

Also, here's something somewhat on point: http://www.eia.doe.gov/emeu/reps/enduse/er01_mid-atl.html

"Typically, an older housing unit will consume more energy for space heating than a newer housing unit. (Although newer homes tend to be larger than older homes, their average energy use per square foot is lower.)

We can guess that much multi-unit housing is much older, and has the old problem of a landlord/tenant split in responsibility. "apartments and condominiums that are individually metered and for which the tenant pays their own heat and hot water use significantly less energy than those that are bulk metered and for which these services are provided at no additional charge -- perhaps in the order of 15 to 25% (e.g., http://www.marketwire.com/press-release/Stratacon-Inc-593630.html). Unmetered consumption can also vary widely by tenant, and it's generally believed that 30 per cent of residents will be responsible for 50 per cent of a building's overall energy requirements." (quoted from http://www.theoildrum.com/user/hereinhalifax).

Probably it's much less efficient by design (single pane windows with R value of .04, very low wall insulation, old furnaces, etc).

What should we do?

Well, it's fairly straightforward. We should stop worrying about whether the suburbs are viable, or whether we should all "localize", and concentrate on the basics: insulation, pluggin air leaks, better windows, higher efficiency lighting and appliances, heat pumps for heating and A/C, and a host of smaller, inexpensive improvements.

Well, that's existing housing. What about new housing?

We nee much, much better "green" design for new homes. For instance, there's every indication that PassivHaus type design can almost eliminate energy consumption for new homes, at very little additional cost.

An Energy Star certified home would reduce energy consumption by 15% over a conventional home. A LEED certified home would reduce consumption significantly depending on the level of LEED certification and the points that the builder chooses to focus on. There are several levels of LEED certification, and the credits can be applied across any number of areas (energy consumption, water reuse, materials resources, etc.).

A DOE ‘zero-energy’ home would reduce consumption by 70%.

The Passive House Standard home is +90% more efficient than a new conventional home, without relying on PV solar arrays and passive solar space heating. See http://campfire.theoildrum.com/node/4946#more

Now, one might ask if energy use per sq ft is the proper was to measure efficiency . IOW, am I suggesting that if SUVs were as efficient on a per pound basis, it would be a good idea to buy one?

No. Here's why:

1) Energy-related design is much more important than building type. Efficient single family homes are available, plug-in SUV's aren't.

2) SUV's and homes are different in another way: homes last 10-20x as long, and we build about 6% as many per year (625K homes per year vs 10M light vehicles). We scrap old SUV's, but we rarely demolish old homes. Even during the white flight of the 50's & 60's, the houses from which people fled weren't abandoned - someone else moved in. If the middle class were to flee the suburbs, the same thing would happen, except the new occupants would have much less money for energy efficiency.

3)Choosing a more efficient vehicle is trivial, while moving homes is a very expensive proposition. Urban condos can be more expensive than suburban homes, even while they're half the sq ft. It's much, much cheaper to make an existing home energy efficient than to move.

4) It's worth noting that different housing types have a different number of occupants: single family homes use less energy per person than townhomes or 2-4 unit apt buildings, and only slightly more than apts. SUV's, on the other hand, don't.

5) SUV's have other problems: they're make the roads much more dangerous, while not improving the safety of their occupants.

6) OTOH, it's true for SUV's as well that design is more important. A plug-in SUV really wouldn't be a big environmental problem (windpower is very cheap to build, should we choose: a one-time payment of $1,500 would pay for the wind capacity needed to power even a very big SUV forever).

The fact is, housing is currently overbuilt. We have more homes than we need, and we're not going to be building a whole lot for a while. We need to primarily pay attention to retrofits.

Energy-related design is much more important than building type. This is true both for existing units, and newly built: conventionally designed existing multi-units aren't more efficient (in fact, they're less), and new single families can be made just as efficient or more.

If you build an apartment building with units having, say, 1800 sq ft with the same techniques as you build a single family home, isn't it true that any of the apartment units are going to consume less energy than the single family home?

Not really - that's too simplistic.

1st, the physics also works against you. You can't put all of the windows on the south side; in a large building you can't use solar for water or PV (some Passive House designs go to net zero energy with solar, which couldn't be done with an apt bldg); you won't have space for ground-based heat pumps (which likely means wasteful resistance heating, or fossil fuels, though air-based heat pumps will be useful in some cases); after a certain point A/C is working mainly to move internally generated heat, not external heat, so no further savings are possible; using earthen berms (or any large inexpensive structural element) for heat buffers will tend to be impossible.

2nd, The Passive House standard sets a maximum, which an architect will design against, so energy consumption will be the same. In particular, exterior windows will be maximized until they hit the limit.

3rd, apts have several common elements that must be taken into consideration. For example, the lighting in hall corridors, stairwells, lobby and other public areas will likely operate twenty-fours hours a day, as will the elevator(s) if the building is so equipped. If there's underground parking, this garage area will be continuously illuminated and perhaps heated during the winter months. There may be any number of ventilation fans that also run 24/7 and even ice melting cables embedded in the sidewalks and parking ramps to keep them free of ice and snow. Most of us would be positively shocked if we were to view the billing records of some of these buildings -- I can tell you I've nearly fallen out of my chair on more than one occasion (this paragraph quoted from http://www.theoildrum.com/user/hereinhalifax).

4th, even if you did manage to reduce further with apts, the absolute difference is going to be so small as to not matter. The Passive House standard is about 3 KWH/SF for heating and cooling, so a 1,800 SF single family will use only 5,400 KWH. If an apt saves 20% beyond that, that's only 960 KWH, or about $100. That could be supplied cleanly by about a onetime expenditure of $700 for wind turbines. For $100 savings we're going to go with apts over another form of housing??

Urban condos can be more expensive than suburban homes, even while they're half the sq ft. It's much, much cheaper to make an existing home energy efficient than to move. I don't know why people waste a lot of thought on very expensive, painful, slow solutions like moving people into different housing, when much cheaper and faster solutions are available.

I've been asked - Shouldn't we replace our housing, to relocalize and improve housing energy efficiency?

No, that would be a very expensive solution. We have very roughly 110M units of residential space. To replace 50%, at a ridiculously conservative $100K each (assuming only 1,000 SF/unit, $100/SF, not including supporting rail or other infrastructure(!) and the demolition of the surplus housing), would cost $5.5B and only eliminate a fraction of our energy consumption (an optimistic estimate would be 50% - your estimate for 12 years out was only 6%). The same money (and sense of political urgency) could build enough wind power to replace all of our coal and gas generation, power all of our transportation and heat pumps for all of our homes, and eliminate 80% of our oil consumption and 90% of CO2.

Didn't we do it before, in the 20 years after 1950?

In 1950 we had an enormous backlog of needed construction - not much happened during the Depression, or WWII. Much of the construction was not replacement, much of the abandoned property was very poor deep rural (some Southern, due to the great migration North). Today we're coming off of a very big construction bubble, with excess property everywhere: suburban and city.

We added much housing, but we didn't lose a big % of the old. We're only building 625K units per year now - to replace as much as 50% of our existing housing in 20 years we'd have to get above 2.75M (a rate we've never reached at peak building rates), figure out how to finance them when residential construction is at the very bottom of anyone's lending list these days, in dense areas that require a great deal more planning and supporting infrastructure (including a lot of rail, most of which would take a decade itself for planning and construction). That's highly unrealistic.

Flight to the suburbs had the wind of cheap, greenfield land at it's back. Flight to the city would have the very big headwind of much more expensive, complicated infill housing in front of it (even if all suburban building codes were made more expensive, or urban codes relaxed). It would also face enormous opposition from the existing residents, who wouldn't be eager to double the density of their neighborhoods and lose many of their homes. It would require the demolition of a lot of existing urban/near urban SFH's and low-rise housing, to be replaced on the same site by townhouses and medium and high rises - this would require local permissions that would be impossible without a WWII-like emergency and no other alternatives, when such alternatives are obviously available, faster and more energy and cost-effective.

So, to summarize: we won't change our settlement patterns and building practices to require less energy so that less dense energy sources like solar and wind suffice?

That will take decades, and provide only marginal gains, relative to what we need.

Building out wind (and solar, etc) would be much, much cheaper and faster: compare 100M residential units at $200K each and only 625K units being built per year, vs 200,000 5MW wind turbines at $10M each.

Rebuilding the residential units would cost $20T and take many decades, while the wind would cost $2T and could be done in 10 years.