Is the end of sprawl in sight?

Eduardo M. Penalver in yesterday's Statesman:

With credit tight and the demand for housing drying up — sales of new homes fell last month to the lowest level in 12 years — new construction in the exurbs is grinding to a halt. The result is a decline in the building industry's appetite for rural land on the urban edge. The question is whether that decline will last. In the past, a sudden drop-off in demand for housing in the exurbs would have represented merely a hiatus. Builders would have bided their time until the housing market recovered, and the outward push would soon have begun again. But persistently high gas prices may mean that the next building boom will take place not at the edges of metropolitan areas but far closer to their cores. When gas costs less than milk, people are more willing to drive 20 miles each way to work every day, burning a couple of gallons of gas in the process. But as gas prices climb, long car commutes become a rising tax on exurban homeownership, and the price people are willing to pay for homes in remote areas will fall.

Increasing gas prices might not be enough to cause people to move, which is why demand for gas proves so inelastic in the short term, but it can influence where people choose to live when they are forced to relocate for other reasons. As the New Urbanist News recently reported, during the present downturn, accompanied as it has been by high gas prices, homes close to urban centers or that have convenient access to transit seem to be holding their value better than houses in car-dependent communities at the urban edge. A recent story in the Minneapolis Star Tribune blamed flagging growth in the Twin Cities' outer suburbs on rising gas prices. If prices at the pump continue to increase, as many analysts expect, the eventual recovery of demand for new housing may not be accompanied by a resumption of America's relentless march into the cornfields.

"Gas prices will kill sprawl" is the conventional wisdom, but I think it is almost certainly wrong.  I don't dispute that rising gas prices will have some effect on the amount of new suburban construction, but I think the effect will be relatively small.  Penalver's dramatic prediction is wishful thinking, not the result of clear-eyed analysis.

Suburbs will continue to be built as long as buyers are willing to pay enough to make it profitable to build them.  Rising gas prices will choke off new construction only if they dramatically reduce the amount buyers are willing to pay. 

A simple hypothetical will illustrate why this won't happen.  Let's suppose Jim is thinking about buying a $200,000 suburban house.  Buying the home would require him to commute 40 miles round trip.  Jim's car gets 20 mpg, so he'll burn 2 gallons per day for his commute -- 10 gallons per week, 500 gallons per year.

The day before Jim makes an offer on the house, the price of gas shoots up $1/gallon.  We'll assume that Jim expects this price increase to be permanent.  The cost of his prospective commute suddenly has shot up by $500 per year.

Let's think things through from Jim's perspective.  Jim says to himself, "If I can get the house for $7,500 less, I will cut my mortgage payments by $500/year and exactly offset my extra commuting costs.  I'll just offer $192,500 for the house rather than $200,000."  The seller, recognizing that all of her prospective buyers face the same trade off, is willing to cut the price.  The home still gets sold, but at a lower price. 

But we've overstated the likely effect of the gas price hike on Jim's willingness to pay.  First, Jim will incur commuting costs regardless.  If his next best option is a house with a 20-mile round-trip commute, the real extra cost to him of the $1 price increase is just $250 per year.  Second, there are things Jim can do fairly easily to reduce his expected commute costs.  For example, he can buy a car that gets 30 mpg, which will reduce the cost of his commute by a third.  All things considered, Jim may be willing to buy the house for just a $2,500 break in price -- or a discount of just 1% off the list price.

Even this may overstate the effect on Jim's willingness to pay.  Close-in housing is now slightly more attractive to Jim, but it's also slightly more attractive for all of the other potential suburban buyers.  This increased demand for close-in housing raises the price of Jim's next best substitute.  Jim is in fact willing to pay even more than $197,500 because of the lack of much better deals closer in.  (All of this is just a long-winded way of saying that Jim's demand for suburban housing may be fairly inelastic.) 

When you add it up, Jim still might be willing to buy the house for $199,000.  The $1 increase in gas has reduced the price he's willing to pay by just one-half of one percent.

This is well below the historical annual average increase in home prices and, I submit, unlikely to deter much new suburban construction.  Even sudden increases of $2 or $3 won't have a dramatic effect on buyer demand. 

This is obviously a highly simplified example.  I've made up the numbers, but I think they are fairly typical.  I have not factored in  non-commuting gas costs -- having to drive everywhere for routine errands -- but I'm skeptical that people save that much gas on these kinds of errands by moving closer in.  People who live close to the city center still drive to the cleaners, grocery store, soccer practice, etc.  They may substitute walking in some instances, but not enough to affect the total transportation budget by a significant amount.

I've also ignored some wrinkles that work in my favor.  The above example assumes a monocentric city in which all suburban dwellers commute to a single city center.  Employment is not really distributed this way anywhere.  Jobs are increasingly located in the 'burbs.  (Austin has an unusually high percentage of jobs downtown if you include UT, but even Austin has tons of suburban employment - Dell, the Arboretum, 360, etc.  Rising gas prices might encourage even more suburban development in Round Rock, Bee Caves, and Cedar Park.)

I've also ignored the fact that most households have two commutes to worry about.  Jim may shorten his commute by moving closer to downtown, but this might lengthen his wife's commute to her suburban job.

When you add it all up, even sharp increases in gas prices will likely have little effect on the amount of new suburban construction.  As I said at the beginning, I will concede that they may discourage the marginal suburban development, but it will take more than CW hand-waving to convince me that the end of sprawl is in sight. 

PS.  I wish the Statesman would "buy local" when shopping for its op-ed commentary.   

Comments

[bittergradstudent]

To be fair, Penalver seems to be talking more about exurban development than he is about suburban development. Certainly, the longer commutes involved there will cause more sensitivity to gas prices, though the example of Los Angeles, with the historically high California gas prices also serve as a solid counterexample that exurban development isn't really all that slowed by rising gas prices.

However, the time cost of hour plus commutes certainly does have an economic cost for the individual, too.

[AC]

I took my 40-mile round trip hypo from Penalver, so I don't think we're talking about different things.

The time cost of commuting is certainly important. It's probably an order of magnitude greater than the cost of gas, and can impose a hard limit on suburban growth (although it can also trigger more job relocation to the subrubs). But I don't see any reason for the cost of fuel to raise the average length of commutes. Commute length will be affected by density (zoning) infrastructure investment.

[Tim]

I was lost at:

"When gas costs less than milk"

obviously not a big milk drinker. Considering my milk is still outpacing gasoline by a minimum of $1.00/gallon.

I think that he's reporting on (and misinterpreting) two trends.

1) Currently there's a trend towards being greener. Some people have noticed that their ginormous houses are far too big and are looking again at smaller houses. Of course the place you would look for a smaller house is in the city center close to amenities. I doubt anyone's looking for a 1000 sq/ft house in Cedar Park.

2) While the real estate market was roaring people who didn't make that much could get into 250-300k homes by selling their old homes for big bucks. Selling to upsize like that is mainly psychological. If you believe the housing market is soft you're not going to make the move. Although from what I'm reading the Austin market might be roaring again, so this might be short lived.

All the new build projects I've seen canceled were shaky projects. They were in locations that might be good in 10 years, but currently are so far off the beaten path that people wouldn't even be able to find them to look at the houses.

[M1EK]

Two quick comments:

#1: We do, in fact, drive a LOT less here in the center-city. We have one car and put less than 10,000 miles on it for a family of four - we have five grocery stores within a mile, for instance; rarely if ever drive more than that to run an errand.

#2: The impact of sprawl-forcing regulations are still an order of magnitude greater than gas costs. IE, supply restrictions on inner-city living make it impossible to provide enough units at a price these people can afford, no matter what gas costs.

[John Goodfriend]

Great article. When I read the original article this was based on I came to the same conclusion. I did not put as much thought and analysis as Mr. Penalver. My conclusion was based on the fact that throught continual and dramatic increases in fuel costs over the years Americans' travel for their jobs and recreation has continued to increase. As you can see from my URL, I believe in healthy and proper use of rural land. When the cost of fuel tempts me to try to save the first things that come to my head are a more efficient vehicle or occasional use of public transportation when I go to the city.

[AC]

M1EK, take a look at the article I cited in my last post. It has an elegant explanation of how zoning regulations force excess consumption of land, imposing disproportionate harm on the low income.

[M1EK]

AC,

Yes, obviously I agree with that - just making the claim that as far as I've observed, the reason gas prices have less of an effect is not because people necessarily demand sprawl even at a higher price, but rather because the price of sprawl-alternatives is so artificially high (which is partly but not completely your argument - maybe 50%). Of course "demand" should read "demand curve", but most people don't get that far.

I still submit that errand drives are a lot less time-and-gas-consuming here than in the burbs - not remotely in the same ballpark as you imply. Even your 1940s suburban area (still mostly suburban) is superior in this regard; and mine blows them out of the water.