Monday, September 10, 2007

Square Feet and Hedonics

Highlights of Annual 2006 Characteristics of New Housing

Please note that the estimates shown here are based on sample surveys and subject to sampling variability as well as nonsampling error.

In 2006:

· The average single-family house completed had 2,469 square feet, 769 more square feet than in 1976.

· 78% of all new single-family homes completed were speculatively-built (house and land are sold together as part of the same transaction), up from 65% in 1986.

· 39% of new single-family homes completed have four or more bedrooms, almost double the rate of just 20 years ago.

· 26% of new single-family homes sold have 3 or more bathrooms, almost triple the rate from 1986.

· Half of all single-family homes were completed in the South region, up 10 percentage points from 1976.

· Approximately 90% of all single-family homes completed have air conditioning!

· Approximately 95% of new single-family homes sold have at least a 1-car garage or carport.

· In the Northeast and Midwest (75%) of the homes completed have a basement, but in the West only 20% have a basement and that drops to 10% in the South.

· Across the country, over half (53%) of all single-family homes sold have at least 1 fireplace.

· 25% of new single-family homes completed have a deck, down from 34% in 1996.

· Almost 70% of all new single-family homes sold use gas as the primary source of heating fuel and approximately 30% use electricity as the primary source.

· Attached single-family homes account for nearly 15% of all new single-family homes sold, up from 11% in 1996.

· Currently vinyl siding is the most common principal exterior material at 30% of new single-family homes sold. In 1996 wood was 22% of the share. It has now reduced to 5% in 2006. Regionally the exterior wall material of preference is: Vinyl - Northeast (86%), and Midwest (67%); Brick - South (41%) and Stucco - West (62%).

· The average sales price of new single-family homes sold (including land) was $305,900. In 1996, the average sales price was $166,400. This is an increase of over 84%!

· The average price per square foot for new single-family homes sold was $91.99, up from $64.38 in 1996. Regionally, it is most expensive to build in the West at $120.66 and least expensive to build in the South at $80.32.

· Over a tenth (12%) of all new single-family homes sold were built on lots of at least 22,000 square feet (approximately a half an acre); this is virtually unchanged from 1986 and 1996.
Historical Census of Housing Tables
Plumbing Facilities
In 1990, only 1 percent of our homes lacked complete plumbing facilities. But, things were much different in 1940, when nearly half lacked complete plumbing. Then, about ten States had rates approaching or exceeding 70 percent. In succeeding decades, the proportion of homes lacking complete plumbing dropped dramatically, falling to about one-third in 1950 and one- sixth in 1960. It is interesting to note the States with the lowest percent- age of such homes in 1940 were higher than Alaska, which topped the 1990 list.

Complete plumbing facilities are defined as hot and cold piped water, a bath- tub or shower, and a flush toilet. In earlier censuses, these facilities must have been for exclusive use of a housing unit's inhabitants; this requirement was dropped in 1990.

This is not exactly scientific as it compounds too many factors and makes assumptions about the distribution of housing stock age but it seems that adjusting for true inflation and hedonics does a lot to explain the "hockey stick."


Sac RE Agent said...

now it's all clearer rob. thanks for the hard work.

oh yeah, first

Rob Dawg said...

Thanks. Wait until I start backing out some of the interesting things this exercise has uncovered as to what things go into housing prices and some of the hidden value that hasn't been correctly valued.

Curious said...


Bah, who am I kidding?

H Simpson said...


That makes sense.

1. when the square footage goes up, the systems to handle it usually are more complex too (eg 1940 bungalos did not have dual heating/AC systems).

I grew up in house with a 30 amp breaker box. My 1st home had a 100 amp service.
My current home has a 200 amp service to handle all the loads we put on the power grid.

2. Code changes have a big impact. I have seen the price of a septic system go from $8k to over $25k in the past 20 years. They are now engineered and made to be larger. How much are fireproof shingles on your LA homes compared to what used to be std 30 years ago?

There was no such thing as smoke detectors 40 years ago. Now every house has to have a system. Ditto CO2 and radon remediation systems.

What about how much stronger they engineer and build your houses for shaking? In the NE, we are now at 2x6 on studs instead of 2x4 and need double or triple glazed windows to keep the cold out. Those code changes cost money to implement.

3.Until the past 10 years, you bought a house and there was some new grass and a couple shrubs. Nowadays every house has 20+ grand of landscaping put into it. Someone has to pay for that.

4. Many towns & cities now shake down the developers to help pay for additional infrastructure like schools. Those new costs are added into the price of the home. The permitting processes have gone from hundreds of dollars to thousands in a few short years.

Not to say the prices are out of wack. Just harder to compare new homes with new homes of 1960..


Peter McFerrin said...

In the NE, we are now at 2x6 on studs instead of 2x4

I've seen the opposite in the Chicago suburbs. Where 2x4 used to be the absolute standard, you're now seeing many tracts built with 2x2 framing, to the point that 2x4 is considered worthy of mention in advertisements. Also, I've heard of a large number of new-build Midwestern houses not having basements, which strikes me as almost criminally negligent in the land of tornadoes. Have Illinois builders figured out how to make a foundation slab not crack when the ground freezes, or do they just not give a rat's ass anymore?

Rob, for one of my courses this semester, I'm reading John Quigley's 2004 Journal of Economic Perspectives article, "Is Housing Unaffordable? Why Isn't It More Affordable?" I skimmed it a couple years back; I don't remember much about it but it did a lot of the type of hedonic analysis used here.

Curious said...

Rawb said:
· Over a tenth (12%) of all new single-family homes sold were built on lots of at least 22,000 square feet (approximately a half an acre); this is virtually unchanged from 1986 and 1996.

Which means slightly under(88%) were sold under 1/2 acre. God damn, can you even imagine a 1/2 acre lot in California, anywhere, that goes for less than a million+?

My head hurts at how out of whack this real estate situation is. Oh well, keep it up and maybe I can retire. :shakes head:

Please, Rawb, where might I find housing nirvana at less than $200K on a more than 1/2 acre lot??

Please tell me that I don't have to move to Texas?

I hate me some Tejas, on personal grounds. Sorry. Arizona, New Mexico(preferred, No se), or Nevada.

bcubbins said...

Isn't the Shiller data already adjusted for quality?

Lou Minatti said...

Isn't the Shiller data already adjusted for quality?

How would he do that? It's apples to oranges. There is no such thing as an equivalent house.

Northern Renter said...

I'm glad I saw your comment above about a future endeavour to separate and quantify the different factors within the cost of housing (yes, I reworded that a bit... hope it reflects the original intent). Without such an analysis, your hedonics may not be very important. For instance, they don't take into account the costs or hedonics of the technology used to build a house. Sure, the wiring system now handles a greater amount of power, but the mains boxes (or whatever they are called) now have a standard size that is bigger than it used to be. Is it much more expensive? Sure houses are bigger, but the machines used to build them are better too. Compare a stapling gun to a guy swinging a hammer. And finally, if the price of land is the major factor affecting housing prices (as the builders sometimes claim), then we would have to consider the hedonics of land. They don't make any more of it, but what land they are making is better and more efficient than it used to be ;-}.


H Simpson said...


How the devil does that support a roof of wet snow with 60 mph blizzard winds howling? The leverage is incredible. Blizzard winds with wet snow can exert the same load forces as a hurricane.

Do they use foam in the walls? Because the heat loss would be insane with fibreglas.

I hear you about basements having grown up with them. I got laughed out of a builder's office in Houston when I went to have a home built 5 years ago. And they have twisters all the time.

Guess they play the law of averages. That and they use double drywall for the central bathroom. It's all good...


Lost Cause said...

Price per square foot is a good metric.

Peripheral Visionary said...

@Curious (from the previous post):

As a Bitter Renter™, my answer on the housing price question is: when/if I buy, I will buy at the price that makes sense for me. I learned a long time ago that the way you buy things is that you decide how much you are willing to spend on it, and then that's what you spend. If that puts you at the upper end of the market, fantastic, if that puts you at the lower end, that's the way it goes.

With respect to the amount that the former owner paid for a property--taking into account interest, time value of money, maintenance, upkeep--I don't know and I don't care. Welcome to capitalism! If the former owner walks away with a bundle of money, more power to them, if they lose money, that's their loss.

What price do the Bitter Renters™ expect? A price we can afford, and that's in line with historic valuations. I will be looking in the 2.5 x earnings to 3 x earnings range, and when I buy it will be something in that range. If it's staringly obvious that better accomodations can be had by renting, and that by buying I would be virtually guaranteeing a short-term loss in value, as is currently the case, than I won't buy.

What do I expect prices to do? I can't say for certain, and economic prognostication is a notoriously difficult art. My stance is that I know what my price range is, and when--or if--housing ever moves into that range, then I will consider buying.

serinitis said...


Are you single or married? The traditional 3xincome is based on a married couple's income. You can still do it if you are single, but be prepared for buying a condo or into a worse neighborhood.

tk said...


I'm with you. The house across the street just sold for $950K. We rent pretty much the same house (ok, our yard isn't as nice) for $1900/month. It's a no brainer here in the Bay Area to rent.

Peripheral Visionary said...

A few additional factors to muddy the waters.

- As mentioned by Northern Renter, modern construction is more efficient with better tools, which should reduce the cost of a home.

- On the flip side, older homes are more likely to have detailing that doesn't show up in any statistics, e.g. crown molding, tiled bathrooms, detailed brickwork, fireplaces, etc.

- Square footage is not fully comparable. Square footage is relatively cheap with modern construction techniques, where ultra-cheap fiberboard and siding and roofing are used as opposed to older materials, such as solid wood flooring and siding and shingled roofs. The biggest cost from sq footage is probably the extra land needed, but also the larger foundation and the additional framing (although modern tract homes use very spare framing with widely spaced studs.) I Am Not A Contractor, but if you specced out a house to be built to 1930s specs, even without modern stuff like LAN wiring and A/C, I have no doubts that it would come in *substantially* more expensive than by building to modern specs.

- Differences in quality are not trivial. My perception is that, while older homes are more difficult to work with, they last longer and age more gracefully. Building a modern home with the quality typical of earlier periods--e.g. more sturdy framing, solid wood siding or brick exterior, fireplaces, solid wood flooring, full basement--is going to be much more expensive, but results in a home that will hold its value longer.

* The overall point that home prices are difficult to compare is a solid one. But I think, overall, modern homes *should* be cheaper than they have been historically. Why should they be any different than automobiles or home electronics or clothes, which have become more affordable relative to salaries? But the evidence is clear that, relative to salaries, the cost of homes has actually increased, which would indicate that something is wrong.

Peripheral Visionary said...


I'm single, but the three times salary is partially because I'm early in my career. I'm relatively new to my firm, and they have a habit of bringing people in at the low end of salary ranges.

And right now, three times my salary is a low end one bedroom condominium, so I'm not looking to buy yet. The median home in my area (D.C.) is running around SIX times median salary, if I recall correctly. But with a tremendous supply of condominiums set to come onto the market, those prices won't last.

serinitis said...


I didn't mean to suggest you buy now. Just for you to be aware of how subconscious assumptions can color what you believe is reasonable for you to get. When I purchased my first home, I paid 4xmy earnings (I am single).

w said...

It seems reasonable for homes to be so much larger than in the past as homes have had to keep up with the people who own them.

bcubbins said...

Isn't the Shiller data already adjusted for quality?

How would he do that? It's apples to oranges. There is no such thing as an equivalent house.

There is if it's the same house. The Case-Shiller index is based on comparing the selling price for a house to the previous selling price for the same house. I don't know how they aggregate all those sales into a single index though.

Some houses will get renovated over time, while others are allowed to fall apart. Some of that can be filtered out by omitting the extreme outlying price changes (both positive and negative), and by omitting sales that were too many years apart. They must have other methods as well.

Rob Dawg said...

Excellent comments all. Random thoughts:

2-2.5x earnings is old fashioned. That from a day when so many other living expenses took up a much larger fraction of the monthly budget. Things like transportation and food are much smaller percentages today thus freeing up more "discretionary" income that could go towards upgrading housing. And wow, has it ever. Then there's the related item of two earner families. The discretionary portion rises even more and the risks of income disruption are reduced.

The enigineer in me is bursting to speak out about tools, methods, materials, prefabrication and the like. An air nailer is what?, a 5x labor multiplier compared to hammers. Engineered floor joists recycle scrap wood and are far better than their predecessors. Complain about the price? Not when three competent carpenters can put up 24 foot spans what used to take a skilled crew 3 times as long. We'll talk more of tthis.

I want to apply hedonics to "square footage." I'm sorry but a modern 3200sf structure wastes so much space it isn't funny. Then they go and negate the benefits of triple pane and R-32 with vaulted foyers. Don't get me started on the waste of not having a basement.

My career as a Bitter Landlordˇ makes it easy for me to call the bottom. Rents for cash flow after taxes and I'll pull the trigger again. No emotion, no need to worry about a growing family or potential job changes. Pure rational investment.

It is the rare CSD (conventional suburban deveopment) homeowner who regrets too large a lot. I'm hoping solar and UHI (Urban Heat Island) and other newer concerns push lot sizes back up.

Special note for readers. Be sure to give Pete McFerrin's blog a look. Lots of interesting stuff. Hey Pete!

One aspect of the increasing housing size is that it will make it a lot easier in the coming years for the displaced class of once temporary home purchasers to increase persons per household. Were to go back to the numbers of just 1995 that translates into nearly 4 million surplus houses. Small demogaphic shifts make big differences.

Legion said...

In for 12000 CFC shares at 17, short.

Rob Dawg said...

Gutsy. I think you are right but I also think the market cap is down so much you are liable to get manipulated out of position.

Legion said...

Thanks for the advice, I bought to cover at 16.20 :-)

rrkap said...


Your correction to the chart is double correcting. The Case Shiller index is NOT an aggregation of average home prices. The index is calculated by looking at "sales pairs" that is, taking a house that sold in the most recent period and looking at deed records to see what the prior sales price was for THAT house. They are specifically attempting to look at the price of a constant quality house, so correcting for increasing housing quality is redundant.

Rob Dawg said...

Your correction to the chart is double correcting.

No it isn't. You correctly describe the C-S methodology but you need to think about the houses built in 1950. For a specific example think Levittowns. How many have the same floorplan and materials as they did then? Still got the asbestos linoleum in the kitchen? The wood frame single pane glass? No garage? Oil burner and tank in the basement? No insulation in the attic? 50 amp service and 1 phone pair connectivity?

Peripheral Visionary said...

@Rob: "2-2.5x earnings is old fashioned."

Hmm. Not sure I'm buying it. And I haven't been buying the "20x P/E is the new multiple for the markets" either. 2.5x earnings is the rule of thumb for getting the 30-year (or better yet, that endangered species, the 15-year) so that the payments don't crush you. Ultimately, I'll look at the size of the payments--wait, don't type that response yet--using a standard fixed-rate money-down mortgage to figure out what I can afford, but I've got a good idea that it's going to be in the 2.5x to 3x range.

"Things like transportation and food are much smaller percentages today thus freeing up more 'discretionary' income that could go towards upgrading housing."

Well, sure. I mean, if you look at the CPI, inflation in necessities has been very low for some time, well, excluding food and fuel from . . . err . . . umm . . .

"Then there's the related item of two earner families."

That, I think, is one of the main causes for the run-up in home prices since the 50's (not so much the recent run-up in prices, which actually saw a lot of single people buying.)

"One aspect of the increasing housing size is that it will make it a lot easier in the coming years for the displaced class of once temporary home purchasers to increase persons per household."

Spot-on comment. One of the most likely results of the current turmoil is very simple: people will combine households to save rent. I'm considering just that, splitting accomodations with a family member to get costs for both of us down.

Parents with adult children: don't be surprised to hear that doorbell ring. ;)

rrkap said...

No it isn't. You correctly describe the C-S methodology but you need to think about the houses built in 1950. For a specific example think Levittowns. How many have the same floorplan and materials as they did then? Still got the asbestos linoleum in the kitchen? The wood frame single pane glass? No garage? Oil burner and tank in the basement? No insulation in the attic? 50 amp service and 1 phone pair connectivity?

I am thinking about them. Most of the ones I can bring to mind (here in the SF Bay Area) have the drawback of having suffered through 50 years of wear and are not nearly equivalent to a new house. The last old house I lived in had electrical systems that were no longer to code, a mix of worn out wood framed single pane windows, wretched aluminum framed single paned windows (can you say condensation, boys and girls) and a couple of new double pane windows and crappy vinyl flooring to replace the (far superior) linoleum/asbestos flooring.

I think it is typical for housing to decline in quality (especially compared to what can be bought for the same price now, even in inflated dollars) as it ages, so if anything, the Case-Schiller methodology understates the run-up in value that has happened.

One piece of evidence that is on my side is the ratio of price to rent, which has gotten really out of whack near me. You should not be able to rent a $700,000 house for $1500/month, but that's what they're going for.

MaxedOutMama said...

Yes, but Rob, the Case Shiller is same house sales within a 2 year period, so any long-term changes in quality are minimal. Therefore I think your last set of changes are without merit, although your point about understated inflation does have merit.

If you want to talk quality, my feeling is that I'd much rather have a house built in the 70's or the 50's. The quality of construction materials used was far superior then. A lot of recent construction is quite shoddy. It's far easier to replace flooring and even run new wiring than to have to jack the foundation up.

Rob Dawg said...

Yes, but Rob, the Case Shiller is same house sales within a 2 year period,

Huh? First I've heard of this. Sorry but that doesn't make sense. There just aren't enough existing home sales within 2 years to make a statistically valid sample.

Son of Brock Landers said...

@rob dawg

thanks for the interesting graphs and work. Maybe I am too much of a cynic, but I get the feeling that Shiller stretche things a wee bit because he has a book to write and services to sell. He wrote an article in Forbes about how the common folk can hedge against oil shocks and other risks, and casually mentioned how his company is creating products which would let everyday people has access to hedging. He patted himself ont he back so hard he faceplanted his keyboard.

This might sound naive, but has anyone considered the effects of different levels of municipal zoning on the effect of housing costs? I know from the town I grew up in zoning laws were more stringent than the neighboring town which caused larger lots, better quality homes and pricier homes. The towns were incredibly similar except for these tiny details that showed up in the houses. Midwestern towns in tornado alley not requiring basements sounds idiotic, and has a considerable impact on the total cost.

Rob Dawg said...

This might sound naive, but has anyone considered the effects of different levels of municipal zoning on the effect of housing costs?

Joel Kotkin, Randal O'Toole and Wendell Cox are prominent among the municipal obstructionist complainers. I am at best a lurker. This is because I'm not willing to jump off the boat just yet.

Locally $50,000 in "fees" is a low estimate. And gee, do you think that encourages smaller footprints?

Let us be clear. Expensive houses are CHEAPER for municipalities. Affordable housing?

Funny Circus Bears said...

I took a long position in CFC today @ $16.52.

TK said...

I'm man enough to admit this conversation is completely lost on me. However Funny Circus Bears...I am conflicted about your decision to go long here, but it's prob. a good move for a session or two.

Legion said...


I went short from 17 down to 16.20 then bought to cover for an intraday trade of 9480. I'm glad I pulled out before the rise at the end..what a bunch of bullshit!

I guess we'll see what happens tomorrow with CFC. I still don't understand why people are throwing money at it, I mean hell, they;ve laready tapped 11 billion, they get another 2 billion from BoA, and now they are looking for "fresh" infusion?

I mean hell, these guys are acting like FB's...hmm, let's just borrow more money! That and the fact that laying off 12000 would only account for a savings of 1 billion? This thing is a turd, and it's already showing itself to be a turd this early in the mortgage meltdown. Sooner or later, they won't be able to borrow anymore just like all the FB's it loaned money too, then see what happens.

Legion said...

oh and good luck tomorrow FCB! :-)

incessant_din said...

Shiller does make an attempt to control for quality, and the depreciation argument is a valid counter to Rob's plumbing hedonics. I think you had it closest with the middle line, which uses the shadowstats inflation methodology. Applying corrections based on hedonics without devaluing them for the corrections already in place (i.e. maybe Shiller went 2/3 of the way that Rob does, so only move 1/3 of the way toward the lowest line).

Here is an excerpt that references the 1890+ chart and the controls for quality:

"Figure 1 shows three long-term series of real home prices. All attempt to control for changing size and quality of homes; all are corrected for inflation in consumer prices."

Another hole in Rob's hedonic adjustment argument is Shiller's accompanying plot of real rents. I'm pretty sure rentals now also have inside plumbing and probably even granite countertops and stainless steel appliances, thanks to Team Flopper.

Lastly, Rob is guilty of an accessibility party foul. Red and green lines with similar intensities and textures. Shiller gets 5 HPI points for that alone.

TK said...


I'm more in your camp though FCB may get a little taste as people stupidly flush their money down the toilet. People love to piss their money away on stupid shit, why not CFC? Hey it's a BARGAIN (snicker).

If CFC is acting like a run-of-the-mill FB, remember the parable of Fliptardian. Debt is not income, and eventually, your creditors will wise up to the fact that you aren't in any position to pay them back.

At that time, and God knows when that will be, let the death spiral begin. I'm wary of short plays in general - I've seen them go obscenely wrong for people. But if ever there was "Safe" shorts these are they. WM, CFC, PHM, DHI, KBH..God I've been watching them for a couple years and ONLY in the last few months has the shit hit the fan/ It ain't over. This is all in my humble, economics-ally challenged, but for God's sakes clear thinking opinion.

Legion said...


Well, I've been hearing a lot of grumbling from the 'shoeshine boy' that real estate is becoming a mess lately, so I know that it is starting to hit the masses. Then again, I did go to Chicago recntly and stayed in a hotel where they were having one of those real estate seminars. I was talking to this woman and I brought up Casey Serin and the mess real estaet is going to be in. She sniffed and said "That's nice, tell someone in a real estaet class that they are making a big mistake" got up and walked away. I hope she bought a house:-)

Legion said...

Hey, I just got an idea for a reality tv show.


They can put anyone with multiple investment properties, liar loans, owner occupied bullshit loans, 100% loans which they walked away from, or multiple foreclosures.

What would the show be about? Who cares, just put them all on an island! The only problem I see is finding land large enough to accomodate all of them. I'm thinking Antarctica...

TK said...

Here's another concept. "Fraudster house" where judges sentence people who committed mortgage fraud to live under house arrest in the homes they've neglected for months and months. Hang out on your jungle of a front lawn, live with the beer bottles and used condoms in your master bedroom, watch the broken window glass in the living room, take a dip in the dark green mosquito pool, enjoy what your neighbors have to say about you. Yes, that would be good TV.

aaron said...

22 years at a travelodge

not a bad deal if you are into one room living arrangements.

Curious said...

I remember a couple of cases like this in NY/SF bay area(?) where judges sentenced slumlords to house arrest in their run down and rat infested properties several years ago.

@PV, Serinitis, & Rob,
Thanks (belatedly) for the input.

@ um..whomever it was who called my previous post a rant...guilty as charged. :blush:

@OGG, I lurve you.

MaxedOutMama said...

Rob - full methodology is given here.

They exclude developer sales, sales less than 6 months apart, and anything for which they can determine substantial physical modification. They also weight by age with the most recent paired sales given preference, and they downweight for big differences from the norm.

I'd be really interested to get your take on possible biases in recent years given their methods. I ran some simulations, and because of the flipping bias I came up with approx 2 years center (in recent years)?

Remember, this thing is only SFR. Stable neighborhoods would be grossly underrepresented using their methods in recent years because of speculative first sales in new developments, if for no other reason.

Another interesting source of bias comes in when property taxes are indexed to the original acquisition date. This causes slower turnover in older neighborhoods. Both Florida and CA have this bias.

Take a look and see what you think. Their data is unquestionably head and shoulders above anything else out there, but that doesn't mean it is sacrosanct.

I agree that the original graph is a splice, and spliced data is not reliable, as poor McIntyre and McKitrick took so much heat for pointing out.

Finally, I am bewildered by the idea of eliminating "neighborhood" effects, since neighborhood effects are an integral part of home valuation.