Greater Greater Washington

Living may actually be cheaper in the region's core

The classic rule of thumb, "drive 'till you qualify," holds that the farther you go from a city center, the cheaper the cost of living. But a new report shows how in the DC area, housing near the core and near transit stations can be cheaper when transportation costs are factored in.


Total H+T costs for the DC region. Image from CNT.

The Office of Planning worked with the Center for Neighborhood Technology to customize their "H+T" housing and transportation index for our region, and to incorporate more recent American Community Survey data as well as Census data.

Along the western Red Line corridor in Montgomery County or the Orange Line in Arlington, for instance, housing is fairly expensive, often exceeding 30% of Area Median Income, the standard threshold for "affordability." After studying metro areas across the nation, with excellent transit infrastructure or none at all, they determined that 15% of AMI was a good threshold for reasonable transportation costs.

Some areas have housing costs below the 30% threshold, but when adding in transportation, the total housing plus transportation exceeds 45% (shown in red below). In other areas, mostly in DC and Arlington, housing costs are higher than 30%, but the low transportation cost pulls the total down below 45% (green):


Change in H+T cost. Red areas: Housing < 30% of AMI but H+T > 45%. Green: Housing > 30% of AMI but H+T < 45%. Image from CNT.

This analysis reinforces two conclusions. First, we need to continue to add housing in the areas with lowest transportation costs, including close to jobs. Not only is it greener, but it's the only way to achieve the least possible H+T cost for households.

Second, when talking about building and maintaining affordable housing, some argue that people should just move out to somewhere like Laurel or Bowie where housing is cheap. It might be, but transportation costs also rise, wiping most most or all of the gains and ultimately creating an outer boundary on how far people can move to actually save money.


H+T burdens by commute distance. Image from Beltway Burden: The Combined Cost of Housing and Transportation in the Greater Washington, DC, Metropolitan Area, Urban Land Institute Terwilliger Center for Workforce Housing, 2009.

The most affordable H+T places are the east side of DC, especially east of the river, and Prince George's County inside the Beltway. That's why it's most critical to add housing capacity to the underutilized Metro stations and other places where, historically, developers and banks have been unwilling to invest.

Support us: Monthly   Yearly   One time
Greatest supporter—$250/year
Greater supporter—$100/year
Great supporter—$50/year
Or pick your own amount: $/year
Greatest supporter—$250
Greater supporter—$100
Great supporter—$50
Supporter—$20
Or pick your own amount: $
Want to contribute by mail or another way? Instructions are here.
Contributions to Greater Greater Washington are not tax deductible.

David Alpert is the founder and editor-in-chief of Greater Greater Washington. He worked as a Product Manager for Google for six years and has lived in the Boston, San Francisco, and New York metro areas in addition to Washington, DC. He now lives with his wife and daughter in Dupont Circle. 

Comments

Add a comment »

Can someone overlay this map with an average income map?

by Jasper on Aug 3, 2011 2:29 pm • linkreport

To add my personal experience: I just gladly took a $4,000 pay cut to change jobs from downtown DC (where my commute took at least 40 mins each way) to Falls Church (where it'll take 10 minutes if I catch all the green lights). I calculated savings in gas and parking and I come out way ahead with the new job.

by Simon on Aug 3, 2011 2:34 pm • linkreport

As always, they way over-estimate car ownership costs.

Depreciation is not cash, sorry. And not everybody drives a new car. I think the average value of a car in Arlington was about $6000.

People in Loudon do not work in DC. I don't see how they factored that in.

GIGO.

by charlie on Aug 3, 2011 2:41 pm • linkreport

What happens when you factor in school quality?

by Geof Gee on Aug 3, 2011 2:42 pm • linkreport

An historical comparison of similar data to the period just before construction of the Metro would be cool. Unlike the Subway, the Metro wasn't designed as much for intra-city transit as for commuters. That's why all the lines radiate from the core to the exurbs and there aren't really cross-town routes.

That was great when most [government] jobs were all in the L'Enfant city and the rest of the region just housed its workers. But that suburb housing/core office relationship doesn't apply now. New office parks around the beltway and development islands around non-core metro station (Ballston, e.g.) mean that people are commuting in new directions: living in the core and working somewhere else in the core or in the suburbs. That reinforces D.A.'s point about needing to increase core density.

But this data probably most clearly demonstrates what happens when people STILL live in the suburbs...but then work IN OTHER suburbs, sometimes on the opposite side of the region. Living in Annandale and working in Rockville means driving much more than just to downtown, and make them more expensive and inconvenient places to live.

by Ronald on Aug 3, 2011 2:43 pm • linkreport

I assume the "transportation cost" from any point on the map is always based on a trip to the city center, and not to a job center somewhere else.

by Lou on Aug 3, 2011 2:46 pm • linkreport

Just skimming through the report my first thought was, what about the number of bedrooms. Shouldn't they have controlled for that in the cost of housing? Or maybe that is assumed to already be reflected in the housing costs.

For instance if you have a family of four then their preference would likely be to have 3 bedrooms. Are the housing costs for a 3 bedroom home (condo/townhome/SFH) similar between the core of DC and the areas outside of it?

by Fitz157 on Aug 3, 2011 2:51 pm • linkreport

@Lou

That seems to e part of it, but they also include variables designed to estimate car ownership/usage and relationship to public transit. Seems like their goal is to answer the question, "How much does it cost to go from a house here to...somewhere else, given available travel options.

They use regression analyses and, well, infer a lot. But short of asking every person it might be as good as they can get.

http://htaindex.cnt.org/method.php

by Ronald on Aug 3, 2011 2:51 pm • linkreport

Hmmm, I'm seeing a number of red flags here. First, as was mentioned above, I don't see a factor for number of bedrooms, which is crucial when comparing housing costs. The central core may be more affordable if a studio or 1 BR is all that you need, but many couples and families are going to be looking for 2, 3 or 4 BR housing units, and those are substantially more expensive in central DC than in, say, Fairfax County.

Secondly, I own a car and for several years commuted to Bethesda from Gaithersburg (I commute on foot now), and my total transportation costs were nowhere near 15% of AMI. The estimated cost of car ownership seems high to me.

Also, I notice that the majority of "red" areas are far outside of the DC core--places like outter Fauquier County, Frederick County, Charles County, etc. where driving is likely the only option and multiple hours per day are spent in the car. But housing becomes more affordable much closer in to the city than those areas--numerous areas in Fairfax, Montgomery and PG counties, along with Alexandria, offer housing that is far more affordable than central DC with transportation costs that do not outweigh the housing savings.

I can rent a 3 BR townhouse in Rockville for what a 1 BR in Dupont would go for; there's simply no way that an increase in transportation costs could outweigh the housing savings one receives from a place like Rockville.

by Ben on Aug 3, 2011 3:16 pm • linkreport

But a new report shows how in the DC area, housing near the core and near transit stations can be cheaper when transportation costs are factored in.

It's meaningless. It's not comparing like to like. Comparing the cost of a small condo in the city to the cost of a spacious house in the suburbs is apples to oranges. Comparing the cost of a transit pass to the cost of running a car is apples to oranges.

by Bertie on Aug 3, 2011 3:19 pm • linkreport

"Comparing the cost of a small condo in the city to the cost of a spacious house in the suburbs is apples to oranges."

Or apples to much larger, similarly priced apples

by Jeff on Aug 3, 2011 3:26 pm • linkreport

I'll throw a DC-centric perspective on this post. It ignores the reality of double income families. It ignores that not everybody works for the feds. It ignores the reality that many non-Washingtonians live in one suburb and work in another, but can not use transit because there is no metro from Ashburn to Rockville. It ignores the cost of private schooling to compensate for living in the second worst school district. And finally, it ignores that many people can not move, because their house is underwater.

by Jasper on Aug 3, 2011 3:27 pm • linkreport

1. Central core is the central city, not just downtown. Granted there is a different set of amenities depending on where you are located. If you are ok with living 3/4 mile from a Metro station, and 5 miles or so to Capitol Hill or Downtown and willing to ride by bike (or use transit), you can buy a house for around $325 to $420 K with 2-3 bedrooms (more if you upgrade the basement) with lots (1/12 to 1/6 of an acre) or not (rowhouses) in Manor Park/Brightwood.

Georgia Ave. isn't that great of an amenity, but the commercial districts in Takoma Park MD and Silver Spring suffice.

2. You make this point:

This analysis reinforces two conclusions. First, we need to continue to add housing in the areas with lowest transportation costs, including close to jobs. Not only is it greener, but it's the only way to achieve the least possible H+T cost for households.

That's key.

But it also means that we have to expand the variety of housing types (apartments, condominiums, alley dwellings/accessory dwelling units) if possible in these areas, within extant neighborhoods, which hasn't been traditionally done, in order to better accommodate more people in ways that don't diminish significantly neighborhood quality of life.

I think it's possible, but our housing policies still aren't set up this way.

3. WRT bedrooms, I don't think this is as much of an issue, as fewer than 20% of households have children. The point isn't to design all housing to meet this particular need, although this is an important need to accommodate in newly constructed housing too.

4. Schools are an issue in terms of retention of households as they add children (although from a straight up financial standpoint it's cheaper to let them decamp to the suburbs, since it costs $15K/year minimum to educate a child) which is why the schools need to be improved. Sadly, the Rhee model is totally f*****--basically fire people and hire younger people. Sadly/2, it hasn't been discredited in favor of the Montgomery County model, which is to provide deep teacher, school, child, and family development and support systems and extra resources for the children/schools (Title 1) who need it. Maybe in a few years we'll get it right.

Even though I no longer favor charters, they do expand options and keep more families in the city as their children reach school age. I see that from families with children in my neighborhood, who don't send kids to the local elementary school.

5. It's not a matter of comparing "like to like" the way Bertie conceptualizes it, the general thrust of the post/study supports the argument in Leinberger's _Option of Urbanism_. It's not a matter of offering housing/options that satisfy everybody, just the 60% of the market willing to live in center cities. The people who want cheap McMansions need to be sent to the suburbs. (But people who want expensive Mansions can be accommodated too, at least in parts of the city.)

by Richard Layman on Aug 3, 2011 3:38 pm • linkreport

As others have pointed out, the fundamental problem with this report is that it ignores variations in the size/quality of housing and transportation between suburbs and core areas. Sorry, but a metro pass is not interchangeable with a car. A car is much more valuable, and people factor that value into their decisions about where to live and how to get around. Ditto for housing. You're likely to get much bigger/better housing in the suburbs for a given amount of money than you can get in the core. If you want to sell people on the idea of living in the city and getting around by public transportation, you're going to have to acknowledge these differences instead of pretending that "housing" and "transportation" are undifferentiated commodities like a pound of apples.

by Wesley on Aug 3, 2011 3:50 pm • linkreport

Nice try, but we're overlooking several realities:

First of all, as pointed out elsewhere, vehicle depreciation is not an out-of-pocket cost.

Also, people keep their vehicles longer; once the loan is paid off, ownership costs significantly decrease.

Also, the atricle takes a simplistic look at the cost of housing. It may be true that a single person or a couple without children can live more in the core - if they are satisfied with smaller living quarters and don't need cars. However, families with children need larger - and more expensive - housing.

and what about those things that can't be quantified - quality of life factors such as elbow room, etc? not to mention the availability of quality public schools and accesible open space.

The findings make some good points, but as many with an agenda often do, they make the mistake of assuming that one size fits all.

by ceefer66 on Aug 3, 2011 3:59 pm • linkreport

@Richard Layman wrote:

It's not a matter of comparing "like to like" the way Bertie conceptualizes it, the general thrust of the post/study supports the argument in Leinberger's _Option of Urbanism_. It's not a matter of offering housing/options that satisfy everybody, just the 60% of the market willing to live in center cities.
----------------------------

Why does the first paragraph of the introduction of the report mention families as a group that should consider the costs of living in outlying areas versus the urban core then? Doesn't family usually mean adult(s) with children?

It doesn't look like the report defines "family," so that's why I'm asking.

by Fitz157 on Aug 3, 2011 4:00 pm • linkreport

It's not a matter of comparing "like to like" the way Bertie conceptualizes it, the general thrust of the post/study supports the argument in Leinberger's _Option of Urbanism_.

Leinberger makes the same error, pretending that a 1-bedroom condo in the city is equivalent to a 3-bedroom house in the suburbs on the grounds that both are "housing." And that a transit pass is equivalent to a car on the grounds that both are "transportation." It is quite obvious to people that they can get much bigger and better housing for their money in the suburbs than in the city and that a car is generally a much faster and more convenient way of getting around than buses and trains. Even if the combined costs of "housing" and "transportation" are higher in the suburbs than in the city, most people are willing to pay those higher costs because the quality and quantity is so much better. The result has been massive suburbanization.

by Bertie on Aug 3, 2011 4:02 pm • linkreport

It may be true that a single person or a couple without children can live more in the core - if they are satisfied with smaller living quarters and don't need cars. However, families with children need larger - and more expensive - housing.

I might add that families who wish to keep horses (or pigs or goats) are also less able to live in the urban core.

by oboe on Aug 3, 2011 4:22 pm • linkreport

The question isn't just a straight housing/transportation comparison but also WHAT KIND of housing. A 700 s.f. one bedroom apartment or condo in Shaw versus an 1800 s.f. 3 bedroom townhouse in the mid-suburbs (say Chantilly, Herndon, or Centreville) both renting conservatively at say around $1500/month denote different values. An urbanist perspective would love the nearby walkable shops, bars, museums, restaurants, etc. as well as the transit and bikeable options and if they work in the city it's even better. However, many more people in this area work in the suburbs than in DC itself. If you work in Northern Virginia you may be used to driving around and as much as you complain about traffic you still don't want to give up the freedom to go wherever you want whenever you want. You also might be starting a family soon and while you know it's possible with effort to obtain a good education for your kids in DC it's much easier and safer to settle in jurisdictions where you are where you know the local schools have a superior national reputation. You might also like the extra space in your closet, having a basement and a backyard, having an extra bedroom to house guests and a living room to entertain them in. You're used to driving to malls like Reston Town Center, Dulles Town Center, Tysons Corner or Fair Oaks and the side shopping centers surrounding them and maybe you even enjoy or prefer that experience to heading downtown or to Dupont Circle, U Street, Adams Morgan, etc. You might like to head into the city occasionally on weekends to play but the thought of living and working there with the crime, crowded conditions, etc. (both real and imagined) is a nightmare.

All this is just to suggest that this isn't just a quantitative analysis. It's also a qualitative analysis based on lifestyle preferences people have. Sprawl wouldn't happen without the demand for it.

by Mike O on Aug 3, 2011 4:27 pm • linkreport

1. I disagree with the title. It is much cheaper to be dead.

2. vehicle depreciation is not an out-of-pocket cost. And that doesn't matter.

3. a car is generally a much faster and more convenient way of getting around than buses and trains.

But buses and trains aren't the only competition. My grocery store is a 3 minute walk from my house. My liquor store is 2 minutes away. My dry cleaner is also 2 minutes away. I have a half dozen restaurants that are a 15 minute walk (or 5 minute CaBi ride) from my house, etc... How long does it take you to drive to the grocery? And I never get stuck in traffic. It's hard to get more convenient than that.

So the advantage is not that transit it faster, but that everything is closer because we don't fill up space with parking lots.

Not only that, but when I ride transit I can read. Or sleep. Or work. That's pretty convenient.

4.most people are willing to pay those higher costs because the quality and quantity is so much better.

That's not entirely clear. Many people do, but that may be because they don't understand the hidden cost of a long commute. Or they're making a bad financial decision for social bias reasons - that happens all the time.

by David C on Aug 3, 2011 4:37 pm • linkreport

"You're likely to get much bigger/better housing in the suburbs for a given amount of money than you can get in the core."

This, of course, assumes we all value a bigger home. Is my house in the core small by American standards? Absolutely. But does it work for me? Yep. I'd be miserable in my brother's house in the suburbs, which costs the same as mine. The commute, the inability to walk anywhere, having to take care of the massive yard, you get the idea.

by Birdie on Aug 3, 2011 4:40 pm • linkreport

I generally agree with the sentiment thatmore important in the choice between suburb and city life than any fairly small differences in cost is values.

However, I think the bigger understated cost of living in the burbs is the cost of needing to move if you change jobs. Many more jobs are accessible from the core, hence there's less of a need to move. Also, what matters more is proximity to your industry than proximity to the core. If you work in biotech, you probably want to live close to 270, etc.

by Falls Church on Aug 3, 2011 4:49 pm • linkreport

And to add to what Birdie wrote, a bigger house is more expensive to maintain. It's more expensive to heat and cool. So that's an added cost of a bigger home.

by David C on Aug 3, 2011 4:52 pm • linkreport

cheaper transportation closer in, for a two adult family, does not mean car free.

It means one car and lower VMT, vs two cars and high VMT. So they still have most of the convenience of car ownership, but at significantly lower cost.

by AWalkerInTheCity on Aug 3, 2011 4:53 pm • linkreport

I would love to see both income and perceived transit figures for the splotches of red in the Four Corners and Forest Glen areas on the map. I moved there several years ago from Virginia because of its proximity to transit and walkable areas, and yet they supposedly have high transportation costs. I find that one tough to believe when all of the job centers are less than 10 miles away, and those places aren't even at the very end of the metro line.

by Joe in SS on Aug 3, 2011 4:54 pm • linkreport

Note CNT does NOT assume all people work in the city center. They estimate total transp costs - including suburb to suburb commute costs, and non commute transport costs - using a regression based on ACTUAL transport costs as the dependent variable, and particular neighborhood charecteristics as independent variables. They then estimate the transport costs for areas where actual transport costs are unknown, but nabe charecteristics are known.

by AWalkerInTheCity on Aug 3, 2011 4:57 pm • linkreport

also CNT is not suggesting that cost is the only element in consumer choices. They have said however that neglecting transportion costs tends to make lower density farther out look cheaper, not only to consumers, but to those who make mortgage loans.

by AWalkerInTheCity on Aug 3, 2011 5:00 pm • linkreport

I also agree this study is pretty bogus. However, even if true, I say "good": living in the suburbs should cost more. If people choose to live in the suburbs, then that's fine, but when you take up more space and public resources then you should pay for it.

The problem is that subsidies have turned this upside-down... Mortgage subsidies help suburban homeowners more than urban renters; new-car tax deductions certainly help those in car-dependent areas; and government expenditures on new suburban roads are lauded by suburbanites as necessary improvements to help them get around their clogged roads while more-urban transit and pedestrian improvements, no matter how many people are served and how small a percentage of the overall budget they represent, are viewed as wasteful spending on a privileged elite. Yet even despite all of the wonders that the overly subsidized suburban lifestyle offers, rents and housing costs per square foot in urban areas far exceed the costs of suburban palaces. There is something seriously wrong with that and we will find that such a lifestyle is ultimately unsustainable.

by Adam L on Aug 3, 2011 5:03 pm • linkreport

What happens when you factor in crime rates?

And I don't mean the wishy-washy "oh insurance will cover that, so just add insurance rates" but actual "this will take six months of my life to recover from physically, and the better part of the decade to get over mentally".

by varun on Aug 3, 2011 5:04 pm • linkreport

I live in Herndon and work in DC, but it would be impossible for my household transportation costs to go down significantly. My husband works in Reston, which means he'd have to drive to work regardless of where we live. I, on the other hand, take a Metro bus (5A) to work. We own two cars, but I have put fewer than 3000 miles on my car in the past year and he has put about 5000 miles on his. If we lived in DC (or even Arlington), he'd be putting at least double the number of miles on his, and while I probably could get rid of my car, I probably would not.

Most of my friends and neighbors who live in the Reston/Herndon area have at least one member of the couple working in Reston/Herndon/Ashburn/Chantilly, and in some cases both. I don't see how these metrics can be correct. If at least one member of the couple works in the same general suburb they live in, moving to the core will not decrease transportation costs.

by Alison on Aug 3, 2011 5:04 pm • linkreport

But buses and trains aren't the only competition. My grocery store is a 3 minute walk from my house. My liquor store is 2 minutes away. My dry cleaner is also 2 minutes away. I have a half dozen restaurants that are a 15 minute walk (or 5 minute CaBi ride) from my house, etc... How long does it take you to drive to the grocery? And I never get stuck in traffic. It's hard to get more convenient than that.

But small local stores generally have higher prices and a smaller selection of products than large supermarkets. And limiting your purchases to what you can carry means more frequent shopping trips. Walking instead of driving also means you're exposed to extreme temperatures and bad weather. And mixing commercial businesses in with residential development generally means more noise and litter and traffic where people live. For these reasons and others, Americans have largely abandoned neighborhoods like yours and moved to car-oriented suburbs, where prices are lower, selections are better, driving and parking are easier, residential neighborhoods are quieter and cleaner, and the quality of life is generally better.

Not only that, but when I ride transit I can read. Or sleep. Or work. That's pretty convenient.

When I drive I can listen to music, or NPR, or an audiobook, or a podcast, or talk on the phone. I'm guaranteed a seat. I have more privacy. I can set the temperature to my liking. I don't have to put up with other people pushing and shoving me. I don't have to stand because all the seats are taken. I don't have to put up with noisy/disruptive/obnoxious behavior by other passengers. I don't have to walk to and from the train station in freezing cold or sweltering heat. I don't have to stand around waiting for the bus or train to arrive. If people thought riding transit was a better deal than driving, they'd ride transit. Overwhelmingly, they choose to drive.

That's not entirely clear. Many people do, but that may be because they don't understand the hidden cost of a long commute. Or they're making a bad financial decision for social bias reasons - that happens all the time.

The average commute by transit takes more than twice as long as the average commute by car, so if anyone doesn't understand the hidden cost of a long commute, it's transit commuters.

by Bertie on Aug 3, 2011 5:07 pm • linkreport

@Bertie

Your arguments are nonsensical. Few people take transit because it's too crowded? Ha! You sir, are not Yogi Berra.

by Adam L on Aug 3, 2011 5:18 pm • linkreport

Ooh here's another good one:
If people thought riding transit was a better deal than driving, they'd ride transit. Overwhelmingly, they choose to drive.

Err, not in places with good transit access they don't.

by MLD on Aug 3, 2011 5:21 pm • linkreport

The analysis simply falls apart at the sniff test. Some of the conclusions may be correct, and others may be correct for a subset of the population, but the analysis seems so simplistic and out of whack as to not provide much ammunition for the conclusion the authors, obviously, presuppose.

The assumption that large segments of the population are acting irrationally (or stupidly) is the height of pseudo-scientific ignorance. It usually means you're not properly weighting the driving factors or you're over-weighting the factors to draw a conclusion that you want.

by ahk on Aug 3, 2011 5:22 pm • linkreport

And limiting your purchases to what you can carry means more frequent shopping trips.

It's called a rolling shopping cart.

Walking instead of driving also means you're exposed to extreme temperatures and bad weather.

Heavens to Betsy. How does one survive such hardships?

And mixing commercial businesses in with residential development generally means more noise and litter and traffic where people live.

Not really.

When I drive I can listen to music, or NPR, or an audiobook, or a podcast, or talk on the phone.

So can I.

If people thought riding transit was a better deal than driving, they'd ride transit.

In DC, more people take transit than drive. Same is true in NYC, Boston, SF, Copenhagen, Berlin, and pretty much any city with good transit.

by David C on Aug 3, 2011 5:25 pm • linkreport

One other problem with this study I just realised. It works off the assumption of the average household size (i.e.: 2.8) for the Metro area. But the fact is (unless you're part of the Donner party), it's highly discontinuous data that they're trying to smooth out. I'd love to see this analysis done again for whole number familial units - 1, 2, 3, 4 people. I think it'll reveal that far from H+T costs going down closer to the core, it'll decrease rapidly the further away you go from the core as family size increases.

by varun on Aug 3, 2011 5:26 pm • linkreport

The problem is that subsidies have turned this upside-down... Mortgage subsidies help suburban homeowners more than urban renters; new-car tax deductions certainly help those in car-dependent areas; and government expenditures on new suburban roads are lauded by suburbanites as necessary improvements to help them get around their clogged roads while more-urban transit and pedestrian improvements, no matter how many people are served and how small a percentage of the overall budget they represent, are viewed as wasteful spending on a privileged elite.

The mortgage interest tax deduction has nothing to do with the location of the home. You'll get the same deduction on a $300,000 hi-rise condo in the city as you will on a $300,000 detached house in the suburbs. If city dwellers more often choose to rent than buy, and thereby forgo the deduction, that's their problem. The deduction is available to them just as it is to everyone else.

As for government spending on roads, it is mostly covered by gasoline taxes and other taxes and fees levied specifically on drivers. Road subsidies amount to about 1 cent per passenger-mile. Transit subsidies, in contrast, are more than 70 cents per passenger-mile. The transit ride for which you pay around $1 costs more like $4 to provide. The other $3 comes from taxpayers. That subsidy constitutes an enormous incentive to use mass transit instead of driving. I've never seen any serious analysis demonstrating that government spending incentivizes driving and suburban lifestyles over mass transit and urbanism.

I have no idea what "new-car tax deductions" refers to.

by Bertie on Aug 3, 2011 5:28 pm • linkreport

2 Senators and a Congressman. Priceless

by TGEOA on Aug 3, 2011 5:31 pm • linkreport

David C says "My grocery store is a 3 minute walk from my house."

Bertie says "But small local stores generally have higher prices and a smaller selection of products than large supermarkets."

David C did not say whether the grocery store was large or small. You assumed it was small.

I, too, live a 3-minute walk away from my "large" grocery store. There are plenty of large grocery stores that people can conveniently walk to in DC in a matter of minutes from their residence.

by JOT on Aug 3, 2011 5:36 pm • linkreport

Bertie, the utility of roads - why you'd be willing to drive yourself for hours - relies on the ability to go door to door, with as little walking as possible. The convenience of cars has much more to do with local streets than with federal-aid highways funded by the gas taxes and tolls.

Instead, local governments pay out for those roads from the general fund, effectively a subsidy. I like having roads and sidewalks like this, but I am paying for roads that I don't use. It is a subsidy.

Furthermore, the federal gas tax is no longer able to fund capital improvements to infrastructure, and is mostly tied up on minimal maintenance of existing roads. The US's infrastructure is seriously undermaintained.

by Neil Flanagan on Aug 3, 2011 5:39 pm • linkreport

In DC, more people take transit than drive. Same is true in NYC, Boston, SF, Copenhagen, Berlin, and pretty much any city with good transit.

As far as I'm aware, New York City is the only city in the United States in which trips by mass transit outnumber trips by private automobile. And that only happens because New York is so dense and hence driving/parking is so difficult. New York's high density is a legacy of its age (as is the high density of the other cities you list). We're not building any more cities like New York. Newer cities, like Los Angeles and Houston and Phoenix, are much less dense and the vast majority of trips in them are made by car. And in the suburbs, where most Americans live, cars are even more dominant.

by Bertie on Aug 3, 2011 5:52 pm • linkreport

Neil Flanagan:

I already said that roads are subsidized. That subsidy is on the order of 1 cent per passenger-mile. Transit is subsidized on the order of 70 cents per passenger-mile. If these subsidies were eliminated, the cost of driving would increase by a tiny amount, but transit fares would triple or quadruple. Which mode of transportation do you think would suffer the greater loss of demand?

by Bertie on Aug 3, 2011 5:59 pm • linkreport

@ Bertie

Here are the numbers:

Highway Trust Fund Revenue: $35B
Federal Road Spending: $42B
State Highway/Road spending: $65B

Vehicle Miles Traveled: 3 trillion

So, that amounts to a subsidy of 2.4 cents per mile driven. However, that's not comparing apples to apples. Roads in urban areas, where public transit exists, are much more expensive to build (and probably to maintain) than roads everywhere else. For example, the 495 HOT lanes will cost $2B to construct for 14 miles of highway (which is perhaps a bad example because the majority of that cost is covered by the contractor, but the point is that urban highways are very expensive to build).

by Falls Church on Aug 3, 2011 6:10 pm • linkreport

D'oh...my analysis above forgets state/local gas tax revenue which is used for roads. That amounted to $34B in 2004.

by Falls Church on Aug 3, 2011 6:25 pm • linkreport

-Re: car depreciation not being a cash cost: Huh? Here's a hint: buy a car, new, for 20k. Use it and then try to sell it in five years. Are you gonna get your 20k back?

-Re: walking to grocery is inconvenient, bad weather etc.
No one said it had to be a small store. Furthermore, walking to many places is much, much healthier for you. Remember that nationwide obesity epidemic we are experiencing?

by H Street Landlord on Aug 3, 2011 6:26 pm • linkreport

You're confusing vehicle-miles with passenger-miles. But it doesn't matter to the point anyway. It doesn't really matter whether it's 1 cent or 2 cents. Or even 10 cents. The point is that mass transit users receive vastly higher subsidies than road users, both in cents per passenger-mile and as a fraction of total costs. The incentives from subsidies overwhelmingly favor transit over driving. And yet people still prefer to drive in overwhelming numbers, because cars are simply much faster, more convenient, more comfortable etc. for the vast majority of trips.

by Bertie on Aug 3, 2011 6:34 pm • linkreport

No one said it had to be a small store.

You can't put a Costco or a SuperTarget within walking distance of most of their customers. Or even a large supermarket. A store that is oriented to customers who access it on foot has a limited customer base and hence a limited size. That's why most neighborhood grocery stores are small, and have correspondingly high prices and limited selections.

by Bertie on Aug 3, 2011 6:50 pm • linkreport

Bertie, can you show me a citation that the 1 cent subsidy includes general fund tax revenues? I've been in this argument before, and usually that number describes non-capital, non-local roads. It often includes state funding for highways, but not for collector and local streets.

It's also worth noting that the numbers you are throwing out are national averages, and may not represent greater Washington's numbers.

by Neil Flanagan on Aug 3, 2011 6:54 pm • linkreport

Limiting ourselves to the paltry costs of car traffic sewerage systems means we ignore the very real costs of

- air pollution and climate change
- time wasted sitting in traffic not working, reading, or spending time with people
- obesity from sitting in traffic, plus accidents & crime from moving in traffic
- fighting trillion dollar wars to keep the price of gasoline low
- the huge inefficiencies of suburban life, hidden by that dirty, cheap Saudi oil

So we’re talking subsidies of what — five, ten dollars per passenger mile? This is your country on a cheap oil addiction, and in deep denial. Those in denial have the overwhelming political might, though, so we’ll run the US into the ground to preserve what 1950s car and house advertisements told us we should value.

by Sydney on Aug 3, 2011 7:00 pm • linkreport

As far as I'm aware, New York City is the only city in the United States in which trips by mass transit outnumber trips by private automobile.

Prepare to be more aware. Mode shares are below

City | Auto share | Non-Auto share
New York | 28.7 | 65.8
Washington | 43.1 | 50.4
Boston | 44.7 | 50.8
San Francisco | 46.4 | 45.1

We're not building any more cities like New York. Newer cities, like Los Angeles and Houston and Phoenix, are much less dense and the vast majority of trips in them are made by car.

And that's not sustainable, which is why each and everyone of those cities is adding light rail and heavy rail.

The mortgage interest tax deduction has nothing to do with the location of the home.

No, but people in cities tend to rent more than people in suburbs. So it is a subsidy that goes primarily to the suburbs.

by David C on Aug 3, 2011 7:16 pm • linkreport

I might add that families who wish to keep horses (or pigs or goats) are also less able to live in the urban core.

Indeed, that's reason #2 why we don't live in the urban core.

by Miriam on Aug 3, 2011 7:48 pm • linkreport

Prepare to be more aware.

The numbers you cite refer to commute trips only. Commutes are only about 15% of all trips. Transit is at its most competitive against cars for commutes, because rush hour is when roads tend to be most congested and when transit services tend to operate most frequently. For the 85% of trips that are not commutes, it is very unlikely that transit accounts a majority of trips in any city except New York.

And that's not sustainable, which is why each and everyone of those cities is adding light rail and heavy rail.

How do you know it's not "sustainable?" As far as I'm aware, there is no heavy rail at all in either Houston or Phoenix, and no plans for it. The light rail lines in both cities provide only a negligible share of transportation. Houston and Phoenix are classic post-war sprawl cities, each covering hundreds of square miles. At a cost of $50-100 million per mile, light rail could not possibly substitute for more than a tiny share of car travel even within the city limits, let alone in the metropolitan area as a whole. Similarly, the rail transit system in Los Angeles provides only a small fraction of transportation in that city. The overwhelming majority of trips are by car.

No, but people in cities tend to rent more than people in suburbs. So it is a subsidy that goes primarily to the suburbs.

As I said, people in cities are free to take advantage of the mortgage interest deduction. If they choose not to, that's their problem.

by Bertie on Aug 3, 2011 8:02 pm • linkreport

@ Alison: +1

by Jasper on Aug 3, 2011 8:20 pm • linkreport

Neil,

I have a plane to catch and don't have time to write a long response to your question. The principal researcher in this area is Mark Delucchi of the Institute of Transportation Studies at UC Davis. Here's a paper by Delucchi that covers road subsidies.
http://pubs.its.ucdavis.edu/publication_detail.php?id=1139

Again, the exact number doesn't matter, because under any serious analysis transit is subsidized at a vastly higher rate than roads.

Delucchi has also studied indirect transportation subsidies in the form of negative externalities (pollution, military spending to protect oil supplies, uninternalized accident costs, etc.). He again found that, taking into account all direct and indirect subsidies, the subsidies to mass transit users vastly exceed the subsidies to motor vehicle users, both in cents per passenger mile and as a fraction of total costs.

by Bertie on Aug 3, 2011 9:05 pm • linkreport

David -

Great article and thanks for bringing attention to this issue. Just because housing + transportation costs are cheaper on the eastern part of the region doesn't mean we should just add more housing on the east to just lower household's housing + transportation costs. The region already has a housing/jobs imbalance with many jobs on the west and lots of housing on the east. It would definitely advance regional goals if we added housing near those underutilized metro stations on the east and actually help those households reduce their housing + transportation costs. However, the region also needs to add a significant number of jobs on the eastern side of the region - preferably near transit. For example, Pr. George's County is already over reliant on residential property taxes and suffers from a lack of a commercial tax base and jobs near households. Adding more housing on the eastern part of the region near transit would definitely help, but jobs near transit are critical part of the solution.

by Josh on Aug 3, 2011 9:43 pm • linkreport

The methodology is flawed in a number of ways, but so are the retorts. There are many good schools in DC and there are many lousy schools in suburban districts, even though the aggregate test scores would suggest one bad, one good, and trust me, Montgy County turns out some woefully poorly prepared (yet arrogant) students--I've taught them at the college level.

Housing sizes are smaller in DC, but also the need for space often is reduced. If you don't need multiple cars, can walk to the supermarket, etc., these factors can mitigate differences in space. At a more individual level, so can the design of space--I have better wall space in my condo than I had in my house in ATL, which was twice as big. Many people don't even use the space they have or really make use of their yards. In the suburbs here, it's rare to see children playing in their yards, people gardening, etc. It's one of the oddities of living in this area.

Car ownership is difficult to quantify because of wide variation in value (even when new) and people buy at different points in a car's life cycle. OTOH, long commutes = large costs and contra Bertie, there are tremendous subsidies of roads; the paper he cites only seems to deal with highways, which represent a fraction of roads and streets. It's another good example of why modeling exercises are flawed as "proof" of anything.

The problem with statistical modeling like this is that it's unclear what happens to the varying distributions of variance and accounting for different lifestyle factors. The savings with giving up a car are probably less as one moves up the income scale. OTOH, selling my car helped me recognize better the marginal cost of car ownership and enabled me to have more choices in buying space.

by Rich on Aug 3, 2011 10:48 pm • linkreport

@David C

Your link shows Bertie is correct- NYC is the only place where mass transit has a higher share than driving. I don't think many people would count walking and biking as mass transit.

Also, I agree with Bertie that the mortgage subsidy does not encourage suburban living. If people wanted to live in the city long term, they'd buy a condo and use the interest deduction to subsidize their mortgage payment. That fact that more people in cities rent, rather than buy, seems to indicate that many people don't want to live in the city long-term.

by Andy R on Aug 3, 2011 11:16 pm • linkreport

For the 85% of trips that are not commutes, it is very unlikely that transit accounts a majority of trips in any city except New York.

Conveniently, you don't have those statistics. Unfortunately, it's not a fair comparison. I take transit to work and back. All my other trips are on foot. A suburban worker might drive to work and then make four more trips that day by car. Thus, more car trips than transit trips. But that's indicative of a failure (complete reliance on a car) not a success.

How do you know it's not "sustainable?"

There isn't enough gasoline. The whole idea of cities like Houston and Phoenix is built on the premise of cheap gasoline. But cheap gasoline is gone. And there isn't enough land for parking. And rivers can't handle all the storm water runoff. And cities can't handle the heat island effect of parking. The only way this is affordable is by not paying for all those negative externalities. But we can't do that anymore.

As far as I'm aware, there is no heavy rail at all in either Houston or Phoenix, and no plans for it.

There's light rail in both. That was my point. And if it is so bad, why are these cities building it? I mean if the car is so awesome, why build light rail systems?

Saying that we built cities for cars and cars only, and now everyone drives and so people must prefer driving is like saying that in some parts of the world they made no investment in plumbing and so everyone drinks water with feces in it so they must like drinking water with feces in it.

by David C on Aug 3, 2011 11:28 pm • linkreport

NYC is the only place where mass transit has a higher share than driving.

Not true. Carpooling is not driving. DC has 36.5% who drive and 37.1% who take transit. But I was wrong about Boston and SF. Still I think that the numbers support my original claim that a lot of people think transit it a better option than driving. And for those who live near transit, it is an overwhelming majority. Y'know who prefers driving over transit? People for whom transit isn't an option.

That fact that more people in cities rent, rather than buy, seems to indicate that many people don't want to live in the city long-term.

Even if that were true, it still means we're subsidizing home ownership over renting. And home owning is more prevalent in the suburbs. Ergo, the subsidy goes more unfairly to the suburbs.

Put another way. If we got rid of the deduction, what would happen to home ownership and renting? The first would likely go down and the latter would likely go up - if not, why even have the deduction? If the price of renting went down relative to owning, and there were more rental units in the city, naturally more people would choose to live in urban rental units (that's basic supply and demand - the marginal customer is choosing ownership over renting because of the deduction and choosing the suburbs over the city because it has more homes for sale).

by David C on Aug 3, 2011 11:47 pm • linkreport

The problem with Bertie's statistics is that they don't account for the billions that were spent creating the nation's road infrastructure in the first place before fuel taxes started to pay into the system. From the accounts I've found, the nation's interstate highway system (let alone local roads) has cost $600 billion more than the revenues from fuel taxes, almost all of which came from the nation's general fund.

Now, however, the country isn't building new highways or any big mass transit projects; the ones we have are substantively complete, which means that the gas tax covers a greater proportion of that maintenance. Since the infrastructure is already built, the costs to maintain roadways is lower but oddly not enough to keep current highways in a state of good repair.

The problem is that our nation's infrastructure is deteriorating and we're not spending nearly enough on roads and transit as we should be to remain competitive. If we really wanted to have an honest comparison between the cost of highways versus mass transit, we would include the actual cost of maintaining the highways we have (as opposed to the funds the Feds and state govs are willing to spend), as well as all the externalties attributed to driving include fuel costs, pollution, national security interests to secure source of foreign oil, as well as the costs of building more expensive utility infrastructure to accommodate sprawling developments.

Something tells me that if all these expenditures and externalties were included in the cost of private automobile transportation that walking, biking, and public transportation would be far more economical.

by Adam L on Aug 4, 2011 2:40 am • linkreport

Martin Wachs, former director of the ITS for Berkeley/UC has a report, through Brookings, that says that roads are subsidized to the tune of 50%.

- http://www.brookings.edu/reports/2003/04transportation_wachs.aspx
- http://www.rand.org/about/people/w/wachs_martin.html

The big problem with this whole discussion is that we are discussing different things. Yes, some people prefer center city life. Others prefer the suburbs. No, living in the center city doesn't make sense if you work in the far suburbs, especially if the location of your job(s) isn't in concert with how the transit system is optimized.

The big thing is that there are tradeoffs. In the core of the center city you need a lot of money to have a big house, compared to the far suburbs. But you don't need a lot of money, comparatively speaking, for smaller places. And maybe the quality of life and other benefits you find there make the tradeoff acceptable.

At the same time, what most of the "antis" in this thread are failing to acknowledge, it appears that we are in the beginnings of a major change in our economy, where energy prices are going to continue to increase, and proximity to major activity centers will matter more, not to mention many changes in the mortgaging financing industry that are going to make big houses and less well placed houses way more expensive if not unobtainable by most people.

Yes, research finds that the avg. house size is shrinking, and the costs (in maintenance, mortgage, utilities) of maintaining large houses is growing.

Yes, urban residents use a lot less energy than suburban counterparts: http://www.flickr.com/photos/rllayman/107880025/

On the other hand, in other places of the city, there are "deals" that compare favorably to the suburbs, but you might not have those immediately proximate amenities (e.g., I was challenged by a prominent DC official who couldn't believe that I am not living in a place like Columbia Heights or the new H St.--hey after 15+ years living in an emerging neighborhood gets tiring).

I could go on and on with this. But the general trends pertain.

The real issue is refocusing on the core of a region, even if not necessarily only the center city as Belmont argues for in _Cities in Full_. In our region that means Arlington, Alexandria, and the portions of PG and MoCo closest to DC.

by Richard Layman on Aug 4, 2011 7:12 am • linkreport

I think something that isn't recognized (at least explicitly) with reports like what the CNT prepared is that individuals and families can have very different interests in their housing choices.

For example I lived in Newport News for 6 years and it never ceased to amaze me how many people that I worked with who commuted 30 minutes to over an hour away from places like Gloucester, Smithfield and Poquoson, where driving to the former 2 required crossing bridges subject to raises in order to allow ships to pass by. There wasn't a great difference in the prices of housing, they simply preferred to live in more rural areas.

by Fitz on Aug 4, 2011 8:11 am • linkreport

They are quoting a price of $975 a month in Transportation costs -- (car+transit) for the average "household" in Arlington. Again, the household size trips you up since a household in Arlington may be a different size than a household elsewhere.

That's close to 12K a year. Again, I call BS.

We know from the county tax revenue that the average car in Arlington has a value of around 6000. The only way to get the 12K year figure is build in deprecation, except at those valuation levels there isn't much depreciation. IN fact, for used cars of that age, there has been appreciation in the last 3 years, but that is another matter.

other issues: their model is based on Massachusetts VMT data from 2005-2007. Gas prices rise, some discretionary trips can drop off.

"The Federal Highway Administration (FHWA), citing Intellichoice’s The Complete Car Cost Guide and Complete Small Truck Guide, reports figures on the cost of owning and operating automobiles, vans and light trucks. These estimates are based on the annual average costs over five years, assuming 70,000 miles driven, and include depreciation, insurance, financing, fuel cost, maintenance, state fees, and repairs. Here, costs are broken out for different types of vehicles (e.g. subcompact) and reported as average per mile costs.
Because the FHWA auto ownership cost estimates break out costs for many different vehicle types, these data were used as the primary source. The U.S. Department of Energy’s Transportation Energy Data Book (2002) was utilized to obtain an approximate fleet mix using the Vehicle Stock and New Sales in the US data. Once the fleet mix was determined, these values were applied to the vehicle types as presented by the FHWA. Costs per mile by auto type were calculated for each cost component using the total cost per mile as well as the percent breakdown between cost components. These per mile costs were then multiplied by the FHWA assumed 14,000 miles annually to obtain a total average fixed cost. This method was used to determine the cost of depreciation, insurance, finance charges, license, registration and taxes (see Table 1).
Total average auto ownership costs were calculated to be $5,068 per auto; this value is applied to the modeled results of average automobiles per household."

by charlie on Aug 4, 2011 9:18 am • linkreport

@varun:

What happens when you factor in crime rates?

You make an excellent point. The researchers definitely should have included crime as an additional "cost" of living in the suburbs:

http://www.virginia.edu/topnews/releases2002/lucy-april-30-2002.html

by oboe on Aug 4, 2011 9:33 am • linkreport

"The assumption that large segments of the population are acting irrationally"

Ive met folks from CNT, and thats NOT their takeaway, afaict. Its that 'affordability' estimates used in making housing policy, evaluating mortgage loans, etc that neglect transp costs are misleading, and tend to bias against higher densities.

by AWalkerInTheCity on Aug 4, 2011 9:35 am • linkreport

@ahk:

The assumption that large segments of the population are acting irrationally (or stupidly) is the height of pseudo-scientific ignorance.

Exactly! People are perfectly rational economic actors. Which is why we don't see things like suburban real estate bubbles, and similar phenomenon. Obviously the fact that there are no government incentives for irrational behavior helps matters.

The dumbest thing I ever heard was Larry Summers' critique of "rational markets" theory (in the wake of the LTCM implosion), which opened with the observation "There are idiots. Look around."

by oboe on Aug 4, 2011 9:37 am • linkreport

@David C:

Saying that we built cities for cars and cars only, and now everyone drives and so people must prefer driving is like saying that in some parts of the world they made no investment in plumbing and so everyone drinks water with feces in it so they must like drinking water with feces in it.

Tighten this up a bit, put it on a coffee mug, and you'll retire in a month.

by oboe on Aug 4, 2011 9:40 am • linkreport

@AWalkerInTheCity wrote:

Ive met folks from CNT, and thats NOT their takeaway, afaict. Its that 'affordability' estimates used in making housing policy, evaluating mortgage loans, etc that neglect transp costs are misleading, and tend to bias against higher densities.
------------------------

Why didn't CNT control for the number of individuals in a household? That dictates the number of bedrooms that a household will desire and as such the affordability. That's why I question their methods when assessing affordability.

by Fitz on Aug 4, 2011 10:01 am • linkreport

At the same time, what most of the "antis" in this thread are failing to acknowledge, it appears that we are in the beginnings of a major change in our economy, where energy prices are going to continue to increase, and proximity to major activity centers will matter more, not to mention many changes in the mortgaging financing industry that are going to make big houses and less well placed houses way more expensive if not unobtainable by most people.

This is an excellent point: for many decades our housing arrangement satisfied folks who really loved living in car-dependent culs-de-sac in the exurbs. There was obviously a not insignificant number of Americans who would've preferred to live in urban environments (since the trope of the couple who "had to move to the 'burbs for the kids' sake" is an incredibly common cliche) but just had to suck it up and adhere to the "standard" way of living.

As cultural and socioeconomic factors change in the coming decades, it's beginning to look like that may be flipped on its head. In two or three decades hence, the default could very well be that most folks with children live in smaller places in highly walkable urban/suburban "nodes" rather than in far-flung exurban developments where you have to have at least two cars.

As poverty increases in the suburbs, that puts downward pressure on infrastructure, including public schools. As cities trend middle-class, schools and infrastructure will improve. Before long, suburban-preferring middle-class families with limited means may find themselves moving *into* the city for the schools, and living in two-bedroom condos in walkable neighborhoods "for the kids".

Imagine if the suburbs consisted of either McLean (only 50% more expensive) or Langley Park. As DC's (and the "streetcar" suburbs') population continued to gentrify, living close-in would certainly start looking more attractive, even to those families who dreamed of owning a 4500 sq ft house with a three-car garage.

Obviously "city" here could mean DC, but also Bethesda, Silver Spring, and a even a more "urbanized" Tyson's or Reston Town Center.

by oboe on Aug 4, 2011 10:02 am • linkreport

@Fitz,

Though it's not quite that simple: socioeconomic factors (including HHI and house size) contribute in determining family size. In other words, the relationship between "house size" and "family size" is not unidirectional.

by oboe on Aug 4, 2011 10:04 am • linkreport

@oboe: the problem is that I remember my parents predicting something very similar in 1974. The age of cheap oil was over! And yet here we still are. Now I, too, believe that this time the age of cheap oil really is over! But it's hard to predict the future accurately, since the future hasn't happened yet.

(Note that I have a similar reaction when people say things like, "The majority of people will always rely on their cars for most trips." Always?)

by Miriam on Aug 4, 2011 10:08 am • linkreport

@oboe,

I understand that it is not unidirectional but the number of bedrooms is going to be a bigger function for a household with children. That's why I'm not so sure that you can compare the affordability of housing without controlling for the number of bedrooms.

by Fitz on Aug 4, 2011 10:16 am • linkreport

@Miriam,

I'm not sure my critique actually relies on the end of the age of cheap oil (Though obviously you shouldn't reject that argument out of hand. We were saved by the deux ex machina of a couple of massive oilfield discoveries in the last couple of decades--we can't always rely on the Hand of God to sweep in and save us).

Regardless of how cheap oil is in the future, population growth is going to increase congestion, which will stress our already over-capacity transportation network. That will further tend to make commutes unbearable. Obviously living in the "hub" won't help folks who live in, say, Leesburg and work in Centreville, but that's not the majority of regional residents.

Furthermore, the fact that predicted change in the past was incorrect doesn't mean that things don't change. All you need to do is look at the census numbers in DC to see that something unprecedented is going on. The city has been transformed in just over a decade. That sort of change tends to be self-amplifying: as more middle-class folks choose the city, it becomes more attractive to middle-class folks, and as more of the region's poor relocate to the suburbs to take advantage of cheap housing, the suburbs become less attractive to the middle-class (and more attractive to the poor).

The changes we've already seen--a refocusing on urban growth patterns in the suburbs, and a corresponding transformation of the city as a primarily middle-class phenomenon--has happened outside of any major changes in energy prices, or a major shift in government incentives for that matter.

If either of those come to pass, we'll likely see a further acceleration of those trends.

by oboe on Aug 4, 2011 10:42 am • linkreport

@DA -- nice article, that on the whole I think describes what most home buyers instinctively figure out. Couple of points:
1. Prices in Deanwood are quite low -- $100-150k buys a 3 bedroom detached house in reasonable shape. Elsewhere EotR houses are similarly inexpensive. Why is this not reflected on the maps? Which leads to
2. Obviously schools and crime are a reason why people are not buying in these areas. As others have pointed out, the cost of schools and crime are difficult to quantify. Maybe an ambitious person one could try to include such.

Regrading cheap oil: the price goes up and down. There have been 4-5 oil "crisis" since the US economy became oil-based. An increase in price prompts more exploration, discovery and/or substitutes, which then eases the supply; the price then goes down. Saudia Arabia has decades of proven reserves. Given that the alternatives cost an order of magnitude more, we aren't anywhere near the end of oil.

by goldfish on Aug 4, 2011 10:51 am • linkreport

I always appreciate the critical thinking on GGW but many have gotten caught up in the DC vs the region vein. Clearly, DC wishes to highlight it's competitive advantages, but the study poses the question: how will economic changes such as rising fuel prices and a credit industry seeing high defaults in the burbs affect both urban and suburban locations? I agree housing families in cost efficient areas with good schools is a major challenge. So what does this mean for the region?

by ArtR on Aug 4, 2011 10:53 am • linkreport

@goldfish:

Saudia Arabia has decades of proven reserves.

Maybe the do, and maybe they don't. It all hinges on what you accept as "proven".

From wiki (yeah, yeah):

Since 1982, the Saudis have withheld their well data and any detailed data on their reserves, giving outside experts no way to verify Saudi claims regarding the overall size of their reserves and output. This has causes some to question the current state of their oil fields. In a study discussed in Matthew Simmons's book Twilight in the Desert, 200 technical papers on Saudi reserves by the Society of Petroleum Engineers were analyzed to reach the conclusion that Saudi Arabia's oil production faces near term decline, and that it will not be able to consistently produce more than 2004 levels.[2] Simmons also argues that the Saudis may have irretrievably damaged their large oil fields by over-pumping salt water into the fields in an effort to maintain the fields' pressure and boost short term oil extraction amounts.

Diplomatic cables leaked during the United States diplomatic cables leak in 2011 revealed that Sadad al Husseini, the ex-head of Saudi Arabia's oil monopoly Aramco, warned the US that the oil reserves in Saudi Arabia might in fact be 40% lower than claimed (300bn barrels).[9][10]

They certainly have a strategic incentive to claim they do, though.

But again, I think in the absence of an oil shock, there's a kind of Malthusian limit that suburban growth is going to bump up against, after which we'll see continuing decline in quality-of-life. I know many folks who live close-in, and pay through the nose for housing because suburban commuter lifestyle is simply not an option.

Their ranks are only going to swell as suburban population increases.

by oboe on Aug 4, 2011 11:43 am • linkreport

Two issues -- schools and crime -- are central to this discussion. At a recent community meeting, Kathy Lanier stated that denser areas around Metro have higher crime. She was speaking about the 14th and U Street area.
The best public elementary schools in the city are in less dense, more expensive areas of the city. Parents in these areas demand quality education but also contribut heavily to ensure it happens. Areas without families don't care about schools so the pressure base is very different in core areas.
It needs to be pointed out that Harriet Tregoning had a significant role in the H+T Report to ensure it came to the right conclusion. Doesn't it strike you as strange that the Planning Director is so involved in Transportation issues. It's all about making sure the basis for fundamental changes to the city can be justified -- even if they are based on flawed facts. We saw right through that David!

by Karl on Aug 4, 2011 12:00 pm • linkreport

"Why didn't CNT control for the number of individuals in a household? That dictates the number of bedrooms that a household will desire and as such the affordability. That's why I question their methods when assessing affordability"

I think thats because standard housing affordability models only look at total housing cost, and income, and do not adjust for household size. CNT is not trying to remodel the concept of housing affordability, but to add a transportation component.

I think the term affordability is leading so some confusion. Housing affordability numbers are NOT guides to where the best deals are, as comments here indicate. They are indicators of where incomes are too low to cover housing costs, an indicator of likely future foreclosure problems, unsustainable prices, etc. Of particular importance in housing FINANCE policy. If Ive got a nabe filled with people whose incomes cant support their costs, Im gonna have to be ready for problems - even if those folks are living well with lots of bedrooms vs other folks squeezed in. Why add transp costs and NOT all costs - because housing affordability models assume that only housing needs to be modeled, cause other costs are not geographically variable, hence simple ratios can be use "rent should be less than X% of income, etc". Transportation costs are VERY much geographically variable, usually in the opposite direction from housing costs, and thats CNT's point. Now if other costs are geographically variable, like say food, that perhaps should be added, but I dont know if the improved accuracy would be worth the effort.

We've seen a HUGE amount of foreclosures - focused esp in the less affluent of the outer suburbs - places like Prince William County. There were lots of reasons for that, but ONE reason may be that borrowers and lenders and guarantors of lendors and so forth did NOT take into account high transportation costs when analyzing housing cost to income ratios for those places. PWC may have better schools than DC, commuting by car may be more pleasant than taking a bus in DC, the average house may be larger, etc, etc - but what good does all that do you when you suddenly have a portfolio of foreclosed homes in your lap?

by AWalkerInTheCity on Aug 4, 2011 12:32 pm • linkreport

Regarding oil prices, there is a lot of room to move up.

Oil in the UK gets near $10/G. That is a bit loose because of exchange rates, but guess what, people still drive. Just less. Average car in the UK get about 9000 miles a year, which according to CNC is about the same as Massachusetts.

There is a lot of discretion driving that can be cut out. And the US market is so large than minor shifts like that mean an extra 1 MB/d kick in.

What I don't know is what that means for china. If US consumers would be dead at $10/G, how do consumers in China survive.

The big effect of higher oil prices is through the supplier chain. Forget about ebay and amazon if oil goes to $10/G. Shipping would be too expensive.

by charlie on Aug 4, 2011 12:39 pm • linkreport

@David C

Re: Mode shares. OK. I'd still call carpooling driving, but I see your point. You and Bertie were both mostly right then.

Re: Mortgage subsidies. Your absolutely right- our tax system subsidizes home ownership. The mortgage interest deduction was designed to do that. I think there are few people that would claim otherwise.

That being said, there are lots of subsidies and lots of tax deductions. Businesses that own rental properties have various types of tax deductions that effectively subsidize rentals. And rent control and affordable housing programs definitely subsidize rentals.

I think its clear that the mortgage interest deduction was intended to encourage home ownership (more accurately, the US kept the mortgage interest deduction, when it eliminated other interest deductions, to encourage home ownership). I don't think that necessarily automatically makes it unfair. Maybe it actually was more along the lines of leveling the playing field for would-be homeowners. It's a pretty substantial deduction, so it quite possibly could be a larger subsidy than what we have on rentals (at least, I'm prepared to assume that it is). Even in that case, it might still make sense from a public policy point of view. One of the arguments I've heard is that homeowners tend to have greater stake in improving the community than transient renters.

I still think that ownership rates in cities and suburbs are pretty telling- in the long-term, people want the extra space and generally betters schools in the suburbs. Housing is also expensive in cities, which shows that many urban areas are still desirable places to live. But since ownership rates are lower in urban areas, people don't seem to want to live there long-term. If we're subsidizing ownership over renting, and you wanted to live in the city long-term, why wouldn't you buy a condo?

How do you interpret ownership rates?

by Andy R on Aug 4, 2011 1:17 pm • linkreport

Sorry if this has already been mentioned. I didn't read all (any) the comments. I also didn't really read the study, so it's possible I'm way off base here. Anyway...

Why just focus on costs and ignore benefits?

Why only include housing and transportation? Aren't there more aspects to "living" somewhere?

by james on Aug 4, 2011 1:32 pm • linkreport

Ok, it looks like I was wrong about the report controlling for household size. It also controls for household income and number of commuters. It shows this on page 33. The report states:

This controlled for any variation in the dependent variables that was a function of household characteristics, leaving the remaining variation a sole function of the built environment.

My only problem with that though is I don't think the built environment is transferable. Doesn't this assume that all households hold the same interests in their preferences? Is a condo in downtown DC going to be an acceptable substitute for a SFH located outside the Beltway?

by Fitz on Aug 4, 2011 1:41 pm • linkreport

@james,

The report probably focused on housing and transportation because the costs of those two aspects can be readily measured and quantified.

Of course one could also argue that CNT is an advocacy group and they're going to present certain information in a manner that is favorable to their causes. Which is fine, I'm not hatin', just sayin'.

by Fitz on Aug 4, 2011 1:46 pm • linkreport

@oboe: It all hinges on what you accept as "proven"
...which rises and falls with the price.

there's a kind of Malthusian limit...
This is where we disagree, because Malthus does not account for human ingenuity. For example, when whale oil ran out a substitute became available. I believe the data is on my side.

...that suburban growth is going to bump up against, after which we'll see continuing decline in quality-of-life.

Besides schooling and crime, what was not considered in this analysis is urban pollution. What was driving a lot of suburban growth in its heyday was smog. The exhaust from any car built in the 60s was terrible -- and a lot of people wanted a cleaner place to raise their family. Likewise, sewage treatment did not exist; that is why there are so few houses near the river. These problems are not so bad these days, so urban living is more attractive.

by goldfish on Aug 4, 2011 2:57 pm • linkreport

@goldfish:

Sorry, I wasn't being clear, I meant Malthusian in the sense that population growth will outstrip the suburban infrastructure (think Snowmaggedon-style commuting nightmares year round); not that it'll outstrip our ability to bring energy resources online.

by oboe on Aug 4, 2011 3:05 pm • linkreport

"Why only include housing and transportation? Aren't there more aspects to "living" somewhere?"

Housing affordability is an established concept - its been a concern of housing and urban policy experts for years, and is used as a measure of RE price sustainability, and a way to evaluate the safety of mortgages. Quality of life concerns do not effect ones ability to pay ones mortgage. Cost concerns that do not vary by geography can be addressed with a housingcost/income ratio. The two big costs that vary by geography are housing and transportation. Historically transportation cost has been neglected in these calculations. CNTs big idea is including them together. There are some other costs that may vary by location, like food for example - but probably harder to measure, and the variation is not as important.

In the report linked above, CNT does seem to focus on implications beyond those for housing finance - but thats still where the concepts come from.

by AWalkerInTheCity on Aug 4, 2011 3:27 pm • linkreport

The article title's use of "CHEAPER" does not connect with the data -- cost as a percentage of income. Cost connects to cheap or not cheap. But there are TWO factors that connect here. Affordability may be measured by cost as a percentage of income. But it is NOT ever a straight measure of cost, price, etc. Can you show me otherwise?

by Tom M. on Aug 4, 2011 4:04 pm • linkreport

This particular study appears to use the average mean income for the region to measure affordability - same across the region, so cheap and affordable are effectively equivalent.

IIUC the CONCEPT was developed to address evaluating individuals mortgages - and the way bank mortgage qual rules drove people further out, where there income could qualify, without noting that transp costs would be higher. In this as in some other papers they have done, CNT applies the same index to areas rather to individuals. Whether that makes sense or not is another question. Im just explaining the history of their H+T index as far as I know it, and why it only includes those two factors.

by AWalkerInTheCity on Aug 4, 2011 4:24 pm • linkreport

One of the arguments I've heard is that homeowners tend to have greater stake in improving the community than transient renters.

A recent study showed that the only advantage was that homeowners spent more time in their garden. We'd be better off giving everyone a $200 annual gardening tax credit.

If we're subsidizing ownership over renting, and you wanted to live in the city long-term, why wouldn't you buy a condo?

Condo fees aren't tax deductible. So while we subsidize condo ownership, it comes with an added kick that isn't subsidized. But, we have seen condo ownership skyrocket lately, so maybe people are buying condos in the city anyway.

by David C on Aug 4, 2011 5:26 pm • linkreport

So you're telling me people are willing to make great sacrifices to ensure their children will live in safe neighborhoods, usually farther from pollution, and go to good schools?

I like urban living but I can't blame anyone for wanting to raise kids in Rockville/etc. instead of the city if they can't afford private school tuition. DCPS is apparently improving, but they have outstanding competition in the area.

by EJ on Aug 4, 2011 11:36 pm • linkreport

In my neighborhood at least, we're seeing more and more white families with children moving in. There's been a united effort in Capitol Hill to improve the schools from preschool up. and it seems to be paying off in attracting people who might have opted for the suburbs instead.

And by the way, I have friends who do live in a condo in the city. They raised two boys (and as teenagers they grew to be quite tall -- they're now in college). It can be done. Instead of a theoretical backyard, they used the parks and museums as the boys' playgrounds. They did send them to private school, but figured it was the payoff of not having to own mini-vans or suvs.

by lou on Aug 5, 2011 12:52 pm • linkreport

Conveniently, you don't have those statistics. Unfortunately, it's not a fair comparison. I take transit to work and back. All my other trips are on foot. A suburban worker might drive to work and then make four more trips that day by car. Thus, more car trips than transit trips. But that's indicative of a failure (complete reliance on a car) not a success.

I have no idea why you think that means it's "not a fair comparsion." Yes, cities that are more oriented to transit also tend to be more oriented to walking. That doesn't alter the fact that the vast majority of cities are neither transit-oriented nor walking-oriented. Nationally, only about 5% of commutes are by transit. And transit's share of non-commute trips is almost certainly much lower than even that small fraction. In terms of total passenger-miles of travel, transit accounts for less than 2%.

There isn't enough gasoline.

How do you know there isn't enough gasoline?

There's light rail in both. That was my point. And if it is so bad, why are these cities building it? I mean if the car is so awesome, why build light rail systems?

And my point in response is that these light rail "systems" account for only a negligible share of transportation in these cities. They do not provide a realistic alternative for more than a tiny fraction of the daily car trips made by people who live in Phoenix or Houston. And no remotely feasible expansion of light rail over the coming decades is going to change that.

Even if that were true, it still means we're subsidizing home ownership over renting.

Because ownership implies a commitment to a community that renting does not. It's much harder to pick and move if you're an owner than a renter, so owners are likely to stay in the community for longer and become involved in its affairs.

by Bertie on Aug 7, 2011 10:41 pm • linkreport

Martin Wachs, former director of the ITS for Berkeley/UC has a report, through Brookings, that says that roads are subsidized to the tune of 50%.

The long-term average is more like 70-80%. But roads are only a small fraction of the costs of driving. Road subsidies are on the order of 1 cent per passenger-mile. The total costs of driving are around 35 cents per passenger-mile. So if road subsidies were eliminated, the cost of driving would increase maybe 3%. Transit, in contrast, is subsidized more than 70 cents per passenger mile. Total transit costs are around $1 per passenger-mile. So if transit subsidies were eliminated, the cost of using transit would increase around 200-300%.

Which do you think would have a bigger effect on demand -- a 3% increase in user costs (a 1 cent per passenger-mile increase for drivers), or a 200% increase in user costs (a 70 cents per passenger-mile increase for transit users)?

You complain about the tiny subsidies to drivers, and ignore the vastly higher subsidies to mass transit users.

by Bertie on Aug 7, 2011 10:54 pm • linkreport

I have no idea why you think that means it's "not a fair comparsion."

Because suburban design creates a need to use a car for every single trip and that inflates the number of car trips. One of the features of transit is that it makes for more walkable neighborhoods and more walking. So to say that you can't consider walking - only transit versus driving - is stacking the deck.

How do you know there isn't enough gasoline?

Because we hit peak oil (per person) 30 years ago. And prices are nearing an inflation-adjusted all time high (higher than it's been since 1918). The era of cheap oil is over.

And my point in response is that these light rail "systems" account for only a negligible share of transportation in these cities.

While fascinating, it is not actually a response to my point which is that the current car-centric system seen in cities that grew up in the 20th century is not sustainable, they know it, and that's why they're investing in transit. For example, transit use for commuting is up 10.7% in LA from 2000-2010, while car use is down 3.6%.

Because ownership implies a commitment to a community that renting does not.

No. It doesn't. "the main positive externality of home investment is the number of well-tended gardens in communities with a larger number of owners."

owners are likely to stay in the community for longer

They are more likely to stay, but they aren't more likely to get involved in local affairs. And staying is not always a good thing. "In equilibrium, unemployment rates for home-owners are higher than for otherwise identical renters." Though that paper states "the impact of home-ownership on aggregate unemployment is quantitatively small" that is not the same as "non-existent". I'm sure Obama would love a quantitatively small reduction in unemployment.

by David C on Aug 7, 2011 11:30 pm • linkreport

So if road subsidies were eliminated, the cost of driving would increase maybe 3%.

You're ignoring all of the other subsidies that exist. Subsidies like minimum parking requirements, the free parking fringe benefit and it's nontaxable status, no sales tax on gasoline in most states, etc... Or the cost of the negative externalities which are paid not by drivers, but everyone.

by David C on Aug 7, 2011 11:37 pm • linkreport

Because suburban design creates a need to use a car for every single trip and that inflates the number of car trips.

So what? Why does that mean it's "not a fair comparison?" I understand that you personally don't like cars and suburbs, but that doesn't mean the comparison is not "fair."

Because we hit peak oil (per person) 30 years ago. And prices are nearing an inflation-adjusted all time high (higher than it's been since 1918). The era of cheap oil is over.

Again, how does that mean there "isn't enough gasoline?" The reason we "hit peak oil," if indeed that's true, is because we are using it more efficiently, not because we have hit a limit on production. Automobile fuel efficiency has increased substantially, and will almost certainly increase even more in the future. The CAFE standards have just been increased to 54 mpg by 2025, and market forces may accelerate that increase. And in addition to the more efficient use of gasoline, we are diversifying automobile hydrocarbon fuels (natural gas, biofuels) and increasingly adding electric power to the automobile fleet (hybrids, plugin hybrids, battery electric vehicles, fuel cell electric vehicles). Electricity can be produced by an even greater range of energy sources. The importance of gasoline as an automobile energy source will probably continue to gradually decline. By the middle of the century, most cars will probably be electric. And if for some reason we do need to substantially increase oil production in the future, there are vast unconventional reserves (shale, etc.) that we have barely begun to tap.

For example, transit use for commuting is up 10.7% in LA from 2000-2010, while car use is down 3.6%.

According to the Census Bureau's Journey To Work data, transit's share of commutes in Los Angeles has been flat for the past 30 years at about 7%. Los Angeles added several hugely expensive rail transit lines over the past few decades, but those rail lines serve only a tiny proportion of the Los Angeles metropolitan area, and most of the riders are people who would otherwise be taking a bus.

No. It doesn't.

Yes it does:

Individuals invest in their local environments by volunteering, getting involved in local government, becoming informed about their political leaders, joining non-professional organizations and even gardening. Homeownership may encourage these investments because homeownership gives individuals an incentive to improve their community and because homeownership creates barriers to mobility. Using the U.S. General Social Survey document that homeowners are more likely to invest in social capital, and a simple instrumental variables strategy suggests that the relationship may be causal. While our results are not conclusive, we find evidence that a large portion of the effect of homeownership on these investments may come from lower mobility rates for homeowners.

by Bertie on Aug 8, 2011 1:08 pm • linkreport

You're ignoring all of the other subsidies that exist. Subsidies like minimum parking requirements, the free parking fringe benefit and it's nontaxable status, no sales tax on gasoline in most states, etc... Or the cost of the negative externalities which are paid not by drivers, but everyone.

But those "subsidies" would at most add a few more cents per passenger-mile to the cost of driving. They still wouldn't begin to approach the enormous direct subsidies provided to mass transit. And, of course, if we are to count additional subsidies to automobile users, we must also count additional subsidies to mass transit users. For example, you complain about lost sales tax revenues from drivers. But mass transit users do not pay sales tax on their transit tickets. And transit agencies do not pay sales tax on the fuel they purchase for buses and trains. Or on the purchase of the buses and trains themselves. They get them at wholesale prices.

However you do the accounting, there's simply no way that drivers are subsidized at a higher rate than mass transit users. The vast majority of the costs of mass transit are paid by people other than the users of mass transit. That enormous subsidy encourages massive overconsumption of transit. But even with that massive incentive to use transit instead of driving, the vast majority of people still choose to drive. That illustrates just how much more benefit people get from driving.

by Bertie on Aug 8, 2011 1:21 pm • linkreport

@Bertie:

The question of whether drivers or transit users are more heavily subsidized is a complicated one. If you want to know why transit users are subsidized, you need only shut down Metro Rail for a couple of days, and examine the impact on the nation's roadways.

Subsidies for transit users *are* subsidies for drivers. (At least in the DC Metro area)

by oboe on Aug 8, 2011 1:34 pm • linkreport

@Bertie
The vast majority of the costs of mass transit are paid by people other than the users of mass transit.
Depends on how you slice it. If you want to look at transit overall across the country, this is true. The reason this is true is because we provide a lot of transit service that only exists to provide basic mobility for people who are too poor to drive an automobile. Transit gets them to work and allows our economy to function. If you have a problem with subsidies like this, then your problem is with the entire entitlement system we have set up in this country, not with transit. Same goes for ADA service.

So let's drill down a bit and look at a place where transit competes with driving: here in DC. The DC metro system is subsidized at a rate of 17 cents per passenger mile. Metro bus is about 97 cents, the two together average out to 34 cents per passenger mile. Fare recovery ratio for Metro is 63% - so contrary to your assertion, riders of metro pay MOST of the costs of running it. Bus recovery ratio is 21%, and averaged out the system recovers 46% of its operating costs through fares. So riders are paying about half the costs.

But even with that massive incentive to use transit instead of driving, the vast majority of people still choose to drive. That illustrates just how much more benefit people get from driving.

Actually, it only indicates that people don't have transit options if you're just going to look at overall numbers. When people have convenient transit options that connect their homes and workplaces they choose them.

Randall O'Toole writes a paper that sounds good, and clearly you've read it. The problem is he compares numbers that are not remotely comparable and then uses those comparisons as the be-all and end-all of "driving vs. transit."

by MLD on Aug 8, 2011 2:08 pm • linkreport

The question of whether drivers or transit users are more heavily subsidized is a complicated one.

No it really isn't. The precise amount of the subsidy for each mode is a complicated question, but the direct public subsidies to mass transit are so enormous (70-80% of total costs, and about 70-80 cents per passenger-mile) that they swamp all other factors.

If you want to know why transit users are subsidized, you need only shut down Metro Rail for a couple of days, and examine the impact on the nation's roadways.

But I wouldn't advocate shutting down Metro rail cold turkey. Land use and transportation patterns in the DC area over the past 30 years have developed in accordance with the growth of Metro rail, so the sudden elimination of Metro rail would be very disruptive. Instead, subsidies could be gradually reduced, giving residents and employers time to adapt. Metro rail has not stopped the suburbanization of the DC area. It has merely slowed it. The most recent Census shows that the outer suburbs have continued to grow much faster than the District and the inner suburbs. Gradually phasing out subsidies to Metro rail would accelerate that trend.

by Bertie on Aug 8, 2011 2:10 pm • linkreport

@Bertie:

No [comparative subsidy] really isn't [complicated].

Excellent! Clearly you've done a bang-up job of convincing yourself.

[T]he sudden elimination of Metro rail would be very disruptive. Instead, subsidies could be gradually reduced, giving residents and employers time to adapt. Metro rail has not stopped the suburbanization of the DC area. It has merely slowed it. The most recent Census shows that the outer suburbs have continued to grow much faster than the District and the inner suburbs. Gradually phasing out subsidies to Metro rail would accelerate that trend.

Funny, I think you and I may agree on something. Of course, I think such a move would lead to a Hell on Earth since there is no "adapt" in your scenario. It would certainly do wonders for the value of my Inside-the-Beltway house, though. You, on the other hand seem to think it would lead to a renaissance of suburban quality of life. Heck, we don't have to speculate where such a "reconfiguration" would leave us; just drive down to the Dumfries-Garrisonville corridor.

Yum.

by oboe on Aug 8, 2011 2:39 pm • linkreport

Why does that mean it's "not a fair comparison?"

Bertie, I don't know how many more times I can say this. But let me try it again.

What we have is a two choice system.

If you choose A, you always have to choose X.

If you choose B, you may choose W, X, Y, or Z.

For people who choose A, X will thus make up 100% of their subsequent choices.

But for people who choose B, let's say, X only makes up 5% of their choices and Y makes up 50%.

comparing Y as a choice to X as a choice will make X appear to be the more popular choice. But that's only because group A can't choose Y. When people have a choice between X and Y though, they overwhelming choose Y. So actually, Y is more popular.

I have an analogy. At rest stops along a turnpike (I forget where), there used to be water fountains. And most people chose to drank water from the water fountain with only a few choosing to buy bottled water. So Coke (who owns Dasani) paid to have the water fountains removed. Now everyone buys bottles water. Does that mean bottled water is more popular?

So, in this case you have a choice to live near transit or far from transit. Living near transit gives you many options for trips, but living far away gives you only one option - car. Comparing transit trips to car trips isn't fair, because the car trip number is inflated by people who have no choice. The fair comparison is to compare choices made by people who actually have a choice.

how does that mean there "isn't enough gasoline?'

When prices go up, it is because demand is exceeding supply. So there isn't enough gasoline to meet current demand at the historical price that made driving such an appealing option. This is why car ownership is down. And this is why vehicle miles driven per capita is down. So what you call "using it more efficiently" I call "abandoning the car."

According to the Census Bureau's Journey To Work data, transit's share of commutes in Los Angeles has been flat for the past 30 years at about 7%

Well from the the census' 2009 ACS, Transit use was 11.3%. So that's a pretty big jump from 7%. Not bad.

those rail lines serve only a tiny proportion of the Los Angeles metropolitan area

And they represent only a tiny investment compared to the investment in roads for the entire LA area. So that seems appropriate.

...suggests that the relationship may be causal. While our results are not conclusive,...may come from...

That does not give me a lot of confidence. Especially since you're quoting the same study by the same person when he determined that the main benefit is well-tended gardens and larger membership in bowling leagues. That does not equal good citizenship.

by David C on Aug 8, 2011 2:43 pm • linkreport

Depends on how you slice it. If you want to look at transit overall across the country, this is true. The reason this is true is because we provide a lot of transit service that only exists to provide basic mobility for people who are too poor to drive an automobile. Transit gets them to work and allows our economy to function. If you have a problem with subsidies like this, then your problem is with the entire entitlement system we have set up in this country, not with transit.

According to the Census Bureau, as of 2009, in only 5 of the 366 U.S. metropolitan areas did transit account for more than 10% of commutes. The share of commutes accounted for by people who both use transit and are too poor to drive an automobile is even smaller. And the share of economic activity accounted for by those workers is even smaller still. If the goal is to provide adequate mobility for the poor, then the proper solution is to provide a means-tested transportation credit and let each recipient use it in whatever way works best for his particular situation. Not to give affluent workers in DC or New York hugely subsidized train tickets from their homes in the suburbs to their well-paid jobs in the city.

So let's drill down a bit and look at a place where transit competes with driving: here in DC. The DC metro system is subsidized at a rate of 17 cents per passenger mile. Metro bus is about 97 cents, the two together average out to 34 cents per passenger mile. Fare recovery ratio for Metro is 63% - so contrary to your assertion, riders of metro pay MOST of the costs of running it. Bus recovery ratio is 21%, and averaged out the system recovers 46% of its operating costs through fares. So riders are paying about half the costs.

I don't know where you're getting your numbers from. According to the National Transit Database, in 2009, the total costs of DC metro rail were about $1.1 billion, of which fares covered about $0.5 billion, or about 45%. The system provided about 1.7 billion passenger-miles of transportation, so the subsidy per passenger-mile was about 35 cents. If you seriously think you can show that DC-area automobile users were subsidized at anything like this rate, I invite you to do so. Metro bus fares covered about $113 million of total costs of $650 million. So the subsidy to bus users was about 83% of total costs. The subsidy per passenger-mile for bus users was about $1.26.

by Bertie on Aug 8, 2011 2:50 pm • linkreport

Let me fix this.

the precise amount of the subsidy for each mode is a complicated question, but the indirect public subsidies to driving are so enormous that they swamp all other factors.

by David C on Aug 8, 2011 2:50 pm • linkreport

Oh, wait, I forgot to respond to this:

The most recent Census shows that the outer suburbs have continued to grow much faster than the District and the inner suburbs. Gradually phasing out subsidies to Metro rail would accelerate that trend.

Actually, what the recent Census shows is the metastasizing growth of the outer suburbs. As wealthier middle-class folks continue to stream into the city, the poor continue to stream into the suburbs. Of course "the suburbs" will continue to grow much faster than the suburbs. (The outer suburbs are also growing faster than Bethesda.) That's because housing opportunities for a more diverse group of people are available out there: the working poor, the non-working poor, the lower-middle class, etc, etc...

We can also see the results of this rapid, unplanned growth on the areas highways every day. Heck, it's been a major driver of the flight of the middle- and upper-middle class back into the inner suburbs and core. The god-awful commutes are so bad they drive anyone with a choice either out of the region or into the center.

I think you're fooling yourself if you think that the loss of Metro is going to lead to a diminished importance of the urban core, though. We hear a lot about suburb-to-suburb commuting, but really that's just a different name for "hot mess".

If Metro were to go away tomorrow (or to be phased out over 25 years to allow suburban jurisdictions to "adapt") the urban core would be *more* relevant, not less. Of course, that won't happen because the folks living in the suburbs understand this, and so defend Metro (rail) funding fiercely.

by oboe on Aug 8, 2011 2:53 pm • linkreport

Man, I wish Wendell Cox (aka Mixner) were here. He would be *loving* this conversation.

by oboe on Aug 8, 2011 3:12 pm • linkreport

@Fitz - you keep mentioning the number of bedrooms as a way to measure size and cost of house.

But it's price per square foot that indicates how expensive housing is, not number of bedrooms.

by Tina on Aug 8, 2011 3:14 pm • linkreport

Funny, I think you and I may agree on something. Of course, I think such a move would lead to a Hell on Earth since there is no "adapt" in your scenario.

Huh? Residents and employers would adapt by shifting even faster to the suburbs. If they had to pay anything like the true cost of getting into the city and getting around the city on mass transit they would have an even bigger incentive to live and work outside the city than they already do. If you personally were willing to pay that cost, fine. But in general, making transit much more expensive to use will make using transit (and transit-oriented land use patterns) much less attractive to people.

by Bertie on Aug 8, 2011 3:22 pm • linkreport

Actually, what the recent Census shows is the metastasizing growth of the outer suburbs. As wealthier middle-class folks continue to stream into the city, the poor continue to stream into the suburbs. Of course "the suburbs" will continue to grow much faster than the suburbs. (The outer suburbs are also growing faster than Bethesda.) That's because housing opportunities for a more diverse group of people are available out there: the working poor, the non-working poor, the lower-middle class, etc, etc...

I haven't seen any evidence of middle-class people "streaming" into the city or poor people "streaming" out of it. The 2010 Census shows that over the most recent decade the outer suburbs of the DC metro area continued to grow much faster than the inner suburbs and the District itself. The suburbs are certainly becoming more socioeconomically diverse, but the vast majority of these new outer suburban residents are probably middle class/middle income range. If you have data to the contrary I'd like to see it.

If Metro were to go away tomorrow (or to be phased out over 25 years to allow suburban jurisdictions to "adapt") the urban core would be *more* relevant, not less.

I have no idea why you think so. The downsizing or elimination would Metro would make it harder to get into and get around in the urban core. That would make it less attractive as a place for people to work, shop and live in.

by Bertie on Aug 8, 2011 3:38 pm • linkreport

The downsizing or elimination would Metro would make it harder to get into and get around in the urban core. That would make it less attractive as a place for people to work, shop and live in.

Jobs aren't going to move out of the city. The Pentagon isn't moving. The Capitol isn't moving. The White House isn't moving etc...There's a reason why jobs are in the city - it's called the network effect. So it will remain just as attractive a place to work because that's where all the jobs will be - just as now.

Because it will be harder to get into the urban core, the urban core will be a more attractive place to live. Better to already be in there then to try and get in.

Because people will now both live and work in the urban core, they will naturally want to shop there.

That's why he thinks so. [Normally I don't presume to know what people think, but this is oboe.]

by David C on Aug 8, 2011 3:44 pm • linkreport

That's because housing opportunities for a more diverse group of people are available out there: the working poor, the non-working poor, the lower-middle class, etc, etc...

I just looked up the Census Bureau's "quickfacts" for Loudon County. This was the fastest-growing suburb of DC between 2000 and 2010. It grew by a whopping 84% over the decade. Arlington County, in contrast, the inner suburb that is endlessly cited as a prime example of the alleged success or attractiveness of urbanism and transit-oriented development, grew by only 9%. In fact, Loudon overtook Arlington in population during the decade. In 2000, Arlington had 20,000 more residents than Loudon. By 2010, Loudon had 100,000 more residents than Arlington.

Nor are the new residents of Loudon "poor" or even "lower-middle class." The median value of owner occupied housing units in Loudon in 2009 was more than $500,000. The median household income was $114,200.

by Bertie on Aug 8, 2011 3:51 pm • linkreport

Jobs aren't going to move out of the city.

You keep making these predictions that are not only unsupported by evidence but that are contradicted by the evidence. Jobs are already moving out of the city. Making it more expensive to get into and around the city would almost certainly accelerate that trend. In its 2009 study of Job Sprawl, the Brookings Institution classified the DC metro area as undergoing "rapid decentralization" of employment. Between 1998 and 2006, the share of metro area jobs within 3 miles of downtown declined by 2.1%. The share of jobs between 3 and 10 miles from downtown declined by 2.5%. The share of jobs 10 miles or more from downtown grew by 4.6%. The growth of telecommuting will also reduce the need for people to live close to their offices. Even if they only telecommute 1 or 2 days a week, that would be a significant reduction in the burden of physical commutes.

by Bertie on Aug 8, 2011 4:10 pm • linkreport

Huh? Residents and employers would adapt by shifting even faster to the suburbs.

This just seems to me a completely unexamined assertion. Let's unpack this for a minute: First of all, there is no such thing as "the suburbs." *Which* suburb will residents and employers shift to "even faster"? There are a million and one points on the map which are all to a lesser or greater degree suburban or exurban.

So which suburb are you talking about? Will all the employment opportunities move to Silver Spring, Tysons, Springfield, Dumfries, and Rockville? How will it be any easier to commute from disparate points around the region to one of those, less-connected end-points? Perhaps a large part of the employment capacity will move to Loudon County. Or maybe we'll just triple the office capacity down in Dumfries. Will we all move there? We'll have to, right, in order to take advantage of the benefits of the easy commute from suburb-to-suburb. What if my wife works in Silver Spring, and I work in Dumfries. Does the fact that we work and live in "the suburbs" make things easier? Doesn't seem to now. I wonder what will happen if we put a few more tens of thousands of private automobiles on the road.

You see the hub-and-spoke model repeat itself throughout nature for a reason. And it's not because Paul Krugman hates the even distribution of objects. It's because that's the way that natural systems tend to arrange themselves.

I'm honestly intrigued: is it your prediction that, should Metro just get defunded, we'll move a couple of hundred thousand more feet of office space to Bailey's Crossroads? That doesn't really seem to address the problem.

by oboe on Aug 8, 2011 4:17 pm • linkreport

Excellent! Clearly you've done a bang-up job of convincing yourself.

I've convinced myself because that is the conclusion overwhelmingly supported by the evidence. Again, Oboe, if you seriously think you can show that motor vehicles users are subsidized at a rate even remotely close to that of mass transit users, then please do so. Remember, transit users are subsidized more than 70% of total costs in direct subsidies alone.

by Bertie on Aug 8, 2011 4:21 pm • linkreport

Jobs are already moving out of the city.

Not really. People are moving out of the city. And then they want to go to TGIF's so they need a TGIF in their neighborhood and so one opens and employees people.

There are more jobs in the suburbs now then there were before, because there are more people than there were before. Jobs are like rats, you'll find them wherever people are. But few jobs picked up from the city and moved to the suburbs. Not over the last decade at least. And the new jobs in the suburbs are generally in the services sector, not the core job sectors (and high salary positions) that define DC (government, law, lobbying etc...)

by David C on Aug 8, 2011 4:26 pm • linkreport

Nor are the new residents of Loudon "poor" or even "lower-middle class." The median value of owner occupied housing units in Loudon in 2009 was more than $500,000. The median household income was $114,200.

Also, the median HHI of Upper Marlboro, MD is $96k. That doesn't refute the growth of poverty in PG County, nor its corrosive effects. We've just demonstrated that suburban demographics are "lumpy".

The city (and inner suburbs) are becoming less so; the mid- and outer suburbs, more so.

by oboe on Aug 8, 2011 4:28 pm • linkreport

So which suburb are you talking about?

The outer suburbs. The ones that are the most car-oriented. The ones that are already growing the fastest. The counties of Loudon, Prince William, Stafford, Prince George's etc.

Does the fact that we work and live in "the suburbs" make things easier?

Given growth and migration patterns, most people seem to find it more attractive than living and working in the city.

I wonder what will happen if we put a few more tens of thousands of private automobiles on the road.

It'll make the suburbs even more attractive than the urban core than they are now, because suburbs are more conducive to getting around by automobile than the urban core.

by Bertie on Aug 8, 2011 4:31 pm • linkreport

Again, Oboe, if you seriously think you can show that motor vehicles users are subsidized at a rate even remotely close to that of mass transit users, then please do so. Remember, transit users are subsidized more than 70% of total costs in direct subsidies alone.

Regardless of the raw numbers, all area residents subsidize mass transit because they realize it's in their best interest. It delays the collapse that poor suburban growth models tend towards. At least for a while. You've as much as conceded this point when you talk about a long period of "weaning off" transit to allow suburban communities to "adapt".

The problem is, as I pointed out previously, there's nothing to adapt to.

by oboe on Aug 8, 2011 4:35 pm • linkreport

Again, Oboe, if you seriously think you can show that motor vehicles users are subsidized at a rate even remotely close to that of mass transit users, then please do so.

Done and done.

"The report divides the external costs of gasoline
usage into five primary areas: (1)Tax Subsidization of
the Oil Industry; (2)Government Program Subsidies;
(3) Protection Costs Involved in Oil Shipment and
Motor Vehicle Services; (4) Environmental, Health,
and Social Costs of Gasoline Usage; and (5) Other
Important Externalities of Motor Vehicle Use.
Together,these external costs total $558.7 billion to
$1.69 trillion per year, which, when added to the retail
price of gasoline, results in a per gallon price of $5.60
to $15.14."

Did I mention that is in 1998 dollars?

by David C on Aug 8, 2011 4:38 pm • linkreport

The outer suburbs. The ones that are the most car-oriented. The ones that are already growing the fastest. The counties of Loudon, Prince William, Stafford, Prince George's etc.

Ok, everybody! Show of hands: Who's excited about your office moving to Dumfries? I want you to visualize the last time you were stuck in traffic coming back from the beach. Now imagine doing that every single weekday for the rest of your working life! Tempting, eh?

What was it Orwell said about a boot stomping on a human face forever?

Heh.

by oboe on Aug 8, 2011 4:42 pm • linkreport

suburbs are more conducive to getting around by automobile than the urban core.

Not in my experience. Not at all.

I spent several years commuting to Rockville (for family reasons). The simplest trip to lunch entailed fighting traffic. I worked down in Dumfries for a brief period several years ago. To go a single exit south, you had to get on I-95 and sit in a traffic jam for 15-20 minutes.

I'm curious, Bertie, do you live around here? I ask because whenever I travel, and meet someone who used to live "in DC", the most bitter transplants are folks who used to live in the suburbs and cite traffic as the singlemost pressing reason they left.

Obviously there may be suburbs where that's not the case currently. But that's because there's no job center nearby.

by oboe on Aug 8, 2011 4:47 pm • linkreport

Oh, and when I need to run to the grocery store or out for lunch, I usually just hop on a bike and ride across Capitol Hill, or down to H Street. But when I don't, I rarely encounter much snarled traffic in the city. Of course, I avoid the downtown area during rush hours.

by oboe on Aug 8, 2011 4:50 pm • linkreport

Bertie wrote: A store that is oriented to customers who access it on foot has a limited customer base and hence a limited size. That's why most neighborhood grocery stores are small, and have correspondingly high prices and limited selections.

I live within 5 min walk of two large grocery stores (major chains). My bf also lives in an urban area, within 5 min walk of one of the largest Giant supermarkets in the area.

The idea that only small stores are near urban centers is a ridiculous and DC is a classic example to refute that point.

by LuvDusty on Aug 8, 2011 4:51 pm • linkreport

suburbs are more conducive to getting around by automobile than the urban core.

Which is why the Beltway, I-66 outside the Beltway, I-270 and I-395 are never congested. It's because they're in the suburbs.

by David C on Aug 8, 2011 4:55 pm • linkreport

Done and done

You keep linking to documents that do not say what you think they say. The "report" you cite doesn't even attempt to compare motor vehicle subsidies to mass transit subsidies, let alone show that motor vehicle subsidies are even remotely close to transit subsidies. Most of the supposed subsidies discussed in the report relate to energy and fuel. You seem to have forgotten that both mass transit and private automobiles use energy and fuel. Most transit is buses. Most buses and some trains run on diesel. If the true cost of diesel fuel for buses and trains is much higher than the price paid by transit agencies, then that is an additional subsidy to mass transit, over and above the already enormous direct subsidies paid by taxpayers.

Again, if you seriously think you can show that motor vehicles users are subsidized at a rate even remotely close to that of mass transit users, then please do so.

by Bertie on Aug 8, 2011 5:06 pm • linkreport

Bertie,

You put the transit subsidy at 70%. I showed that the subsidy for driving (mostly relating to fuel, but not entirely) is between 300% and 750%. As I see it, the ball is in your court.

If you seriously think you can show that mass transit users are subsidized at a rate even remotely close to that of motor vehicles users, then please do so.

by David C on Aug 8, 2011 5:11 pm • linkreport

And, I'll add, that there is another subsidy every driver pays - and that is the time they spend driving. We've got people putting in another 0.5-2 hours of work a day (and driving is work, a lot of people do it for a living even) for which they don't get paid. And more if you count all the trips they make by car. Ask most drivers if they'd be willing to pay $3 a day to have someone else do the driving, and I bet they'd say yes. So the dirty secret is by inducing everyone to drive, we're pushing work on to them, which is like a little bitty tax.

by David C on Aug 8, 2011 5:19 pm • linkreport

Regardless of the raw numbers, all area residents subsidize mass transit because they realize it's in their best interest.

In that case, all area residents subsidize automobiles because they realize it's in their best interest. So stop complaining about automobile subsidies. You can't have it both ways. But I don't agree that all area residents believe it's in their best interest to subsidize mass transit. Or automobiles. How can you possibly claim to know what "all area residents" believe. It's absurd.

It delays the collapse that poor suburban growth models tend towards.

I'm not sure what "poor suburban growth models" is supposed to mean. Whatever they are, you haven't produced any evidence that they "tend towards collapse" or are "unsustainable."

by Bertie on Aug 8, 2011 5:22 pm • linkreport

@David C -interesting research report. Probably because it was written in 1998 it left out an important component of population health costs associated with car-centric living and designs that have been more intensly studied in the last 13 years: the obesity epidemic. I think another commenter upthread mentioned it.

Here's one example: http://www.ajpmonline.org/article/S0749-3797(07)00110-9/abstract

by Tina on Aug 8, 2011 5:30 pm • linkreport

Hmm. You must have me confused with someone else. I'm not complaining about automobile subsidies. While it would be nice to use that money for something productive, that's clearly not happening. Instead we use it to degrade the suburban environment and it's returned in a corresponding premium in urban and close-in housing prices.

That's why per square foot urban housing is so much more expensive than that in the exurbs. Also why when the area housing market cratered recently, the worst loses were concentrated in the furthest sprawl-zones. I think you mentioned them by name:

The outer suburbs. The ones that are the most car-oriented. The ones that are already growing the fastest. The counties of Loudon, Prince William, Stafford, Prince George's etc.

From a purely selfish point of view, I prefer policies that make the suburban environment comparatively worse. As sprawl and congestion have gotten ever worse, my urban row-house has quintupled in value. It held that value as foreclosures ravaged the exurbs.

On the other hand, I'm a decent liberal fellow, so I do support the idea of a Greater Greater Washington. Just quite pessimistic about its ever coming about.

by oboe on Aug 8, 2011 5:31 pm • linkreport

You put the transit subsidy at 70%. I showed that the subsidy for driving (mostly relating to fuel, but not entirely) is between 300% and 750%.

You need to read more carefully. Transit is subsidized more than 70% in direct subsidies alone. That is, money collected from taxpayers to pay the direct costs of building rail track and stations, buying buses and trains, buying fuel, paying bus and train drivers and mechanics, etc. If there are hidden costs to fuel and energy that are not reflected in the price transit agencies pay, then that is an ADDITIONAL subsidy to transit, over and above the already enormous direct subsidies. You haven't produced any comparison between automobile subsidies and transit subsidies.

by Bertie on Aug 8, 2011 5:32 pm • linkreport

And, I'll add, that there is another subsidy every driver pays - and that is the time they spend driving.

That's not a subsidy. It's a cost paid by the driver himself. You don't seem to understand what "subsidy" means.

But your point wouldn't make sense even if it were a subsidy. In general, cars are much faster than mass transit. Cars save time compared to transit. The average commute by mass transit takes more than twice as long as the average commute by car.

by Bertie on Aug 8, 2011 5:39 pm • linkreport

Bertie, lets keep track of your central thesis here:

1. It is quite obvious to people that they can get much bigger and better housing for their money in the suburbs than in the city
2. It is quite obvious to people that a car is generally a much faster and more convenient way of getting around than buses and trains.
3. Even if the combined costs of "housing" and "transportation" are higher in the suburbs than in the city, most people are willing to pay those higher costs because the quality and quantity is so much better.
4. The result has been massive suburbanization.

To some extent, 1 is true, but it all depends on how you define better. But the point of the original post is that people are ignoring the added cost of transportation that living far from work adds.

2 is patently false. When people have a choice, they choose trains over driving. And it ignores biking and walking, which people often choose over driving. Living in the city, with everything within walking distance, is much faster and convenient in the minds of enough people that your statement can not be labeled as "quite obvious to people." It is very much in contention.

For 3,you are correct that people are willing to pay that cost, but you fail to prove why. I assert it is because they are ignoring the cost of transportation.

4 falls apart since 2 and 3 are wrong.

by David C on Aug 8, 2011 5:40 pm • linkreport

Not in my experience. Not at all.

Then your experience is bizarre. Congestion is generally worse in urban cores than in suburbs. Congestion increases with urban density, because there are more vehicles per lane-mile of roadway. Parking is also generally more expensive and hard to find in urban cores than in suburbs. Frankly, I'm amazed you don't know this.

by Bertie on Aug 8, 2011 5:46 pm • linkreport

That's not a subsidy.

You're right. I was sloppy. It is an added cost. I know what a subsidy is, but thanks for talking down to me. I really enjoy that.

As to your second point. Even when driving is faster than transit, an hour spent driving is not the same as an hour spent riding as a passenger. In fact studies show that drivers show up to work significantly more stressed than the passengers of cars do. All others things being equal, most people would pay more to take the train to NYC than to drive themselves.

So the point is that an hour of driving is more like work than an hour of riding the bus.

by David C on Aug 8, 2011 5:46 pm • linkreport

then that is an ADDITIONAL subsidy to transit, over and above the already enormous direct subsidies.

And that is how much?

You haven't produced any comparison between automobile subsidies and transit subsidies.

Nor have you.

by David C on Aug 8, 2011 5:48 pm • linkreport

Congestion is generally worse in urban cores than in suburbs.

In the Washington area, all of the most congested roads are outside of DC.

by David C on Aug 8, 2011 5:49 pm • linkreport

To some extent, 1 is true, but it all depends on how you define better.

Housing in the suburbs is generally cheaper (lower price per square foot) and newer (better design, better plumbing, better electrical, better appliances, etc.) than housing in urban cores. Most people seem to think that cheaper and newer is better.

But the point of the original post is that people are ignoring the added cost of transportation that living far from work adds.

You just keep making up facts out of thin air. How do you know people are "ignoring the added cost of transportation that living far from work adds?" What evidence do you have that people are "ignoring" this cost?

2 is patently false.

No, it is not false. The evidence overwhelmingly shows that cars are generally much faster than mass transit. The average commute by transit takes more than twice as long as the average commute by automobile.

And it ignores biking and walking, which people often choose over driving.

For the vast majority of trips, biking and walking are even slower than transit.

by Bertie on Aug 8, 2011 6:01 pm • linkreport

And that is how much [in additional subsidies to mass transit]?

According to your quote, $5.60 to $15.14 per gallon of gasoline. Most transit buses run on diesel. You didn't provide any figures per gallon of diesel fuel, although they would presumably be similar given that gasoline and diesel are both refined from petroleum. You also didn't provide any figures for the fuels used to generate the electricity used by electrically-powered transit vehicles (coal, nuclear, natural gas, etc). To add this indirect subsidy to the direct subsidy, you would need to convert it to a cost per passenger-mile.

The point, which I'm still not sure you understand, is that both automobiles and transit use fossil fuels. If the negative externalities of fossil fuels (pollution, etc.) are to be counted as subsidies, then they are ADDITIONAL subsidies to transit, over and above the 70+ cents per passenger-mile in direct subsidies that I have already documented.

by Bertie on Aug 8, 2011 6:18 pm • linkreport

Nor have you.

Yes I have. Direct subsidies to automobile users are on the order of 1 cent per passenger-mile. Direct subsidies to mass transit users are on the order of 70 cents per p-m. That is a huge disparity. And quibbles about indirect subsidies (parking, pollution, "oil wars", etc.) won't make it go away. Under any remotely serious accounting of costs, mass transit is subsidized at a vastly higher rate than automobiles.

by Bertie on Aug 8, 2011 6:27 pm • linkreport

That's why per square foot urban housing is so much more expensive than that in the exurbs.

No, urban housing is more expensive because it costs more to supply. Land costs are higher because more people are competing for each square foot. Construction costs are higher because it costs more to build vertically than horizontally. In areas where mixed-use residential/commercial development is permitted, homebuyers are also competing for land with commercial businesses that can afford to pay more, which further increases housing prices.

Also why when the area housing market cratered recently, the worst loses were concentrated in the furthest sprawl-zones. I think you mentioned them by name The outer suburbs. The ones that are the most car-oriented. The ones that are already growing the fastest. The counties of Loudon, Prince William, Stafford, Prince George's etc. From a purely selfish point of view, I prefer policies that make the suburban environment comparatively worse. As sprawl and congestion have gotten ever worse, my urban row-house has quintupled in value. It held that value as foreclosures ravaged the exurbs.

Your narrative doesn't make any sense. Loudon County grew almost ten times as fast as Arlington County. Prince William County grew more than four times as fast as Arlington. Stafford County, almost four times as fast. Whatever effect foreclosures have had, they clearly have not "ravaged" the suburbs or exurbs. The population distribution is shifting AWAY from DC and the inner suburbs.

by Bertie on Aug 8, 2011 6:53 pm • linkreport

@David C

Congestion is generally worse in urban cores than in suburbs.
In the Washington area, all of the most congested roads are outside of DC.

This is what makes me believe @Bertie must live in suburban Cleveland, or maybe Detroit. There really is a difference between the 5-6 megalopolii and the rest of "urban" America.

No one who's lived in the DC Metro area could hold to the idea that "the city" is congested while "the suburbs" are a model of efficient traffic flow.

by oboe on Aug 8, 2011 6:54 pm • linkreport

This is what makes me believe @Bertie must live in suburban Cleveland

And this is what makes me think you have a habit of formulating your beliefs through wishful thinking rather than looking at the evidence. From Sprawl and Urban Growth by Ed Glaeser:

While congestion may be slowing commutes in some edge cities, in most cases, the car-based urban frontier has far shorter commutes than the old public transportation cities (Gordon, Kumar and Richardson 1991). Cars are just so much faster than public transportation that commutes in the old dense cities are almost always much longer than in their newer, sprawling competitors.

In Table 9, we use the micro data from the 1995 National Personal Transportation Survey to study differences in one way commuting times (in minutes) as a function of distance to work and residential block density. We find that average commute times rise with population density. The effect of density is actually less on car commuters than on noncar commuters. It is also true that across cities, there is a strong positive relationship between average commute times and the logarithm of population density

by Bertie on Aug 8, 2011 7:28 pm • linkreport

Your narrative doesn't make any sense. Loudon County grew almost ten times as fast as Arlington County. Prince William County grew more than four times as fast as Arlington. Stafford County, almost four times as fast. Whatever effect foreclosures have had, they clearly have not "ravaged" the suburbs or exurbs. The population distribution is shifting AWAY from DC and the inner suburbs.

I think the point you're missing is that there are many, many reasons why "population" might grow. We're not in a competition for souls. It's great that the racist housing policies of the past are largely forgotten. It's great that suburban constituencies are finally starting to provide services for the very poor. There are middle-class people who want to live in urban environments. There are middle-class people who just love the suburbs. There are also lots of middle-class people who would prefer to live in the city, but sadly can't because "the schools are bad."

As the legacy of Jim Crow fades in the rear-view mirror, it'll become more and more apparent what the preferences of Americans are. But price per square foot is a pretty good indicator. It holds in the case of McLean versus Bowie. It holds in the case of Dupont Circle versus Centerville.

by oboe on Aug 8, 2011 7:38 pm • linkreport

Loudon County grew almost ten times as fast as Arlington County.

At any given Nationals game, there are more people sitting in the far outfield bleachers than sitting behind home plate. Obviously people prefer to sit 500 yards away in the bleachers.

by oboe on Aug 8, 2011 7:40 pm • linkreport

@Bertie:

While congestion may be slowing commutes in some edge cities, in most cases...

I'll keep that in mind if my wife and I ever move to "Most Cases, USA". As it is, we're living in the greater Washington, DC metro area, where this observation certainly does not hold.

by oboe on Aug 8, 2011 7:43 pm • linkreport

The trend of the past 50 years or more, not only in the United States but in virtually every other wealthy democracy, of increasing suburbanization and an increasingly car-oriented transportation system constitutes rather overwhelming evidence that the preferences of most consumers and voters is for car-oriented suburban lifestyles, not transit-oriented urban ones.

And your claim about prices as an indicator of demand is just wrong as a matter of basic economics. You cannot infer anything about relative demand from price comparisons alone. You also have to account for the cost of supply. The price of truffles is higher than the price of potatoes not because there is more demand for truffles than potatoes but because truffles cost more to supply. The price of dense urban housing is higher than the price of suburban housing because dense urban housing costs more to supply. If people wanted lots more dense urban housing their choices as voters and consumers would reflect that preference.

by Bertie on Aug 8, 2011 7:57 pm • linkreport

At any given Nationals game, there are more people sitting in the far outfield bleachers than sitting behind home plate. Obviously people prefer to sit 500 yards away in the bleachers.

I don't know what point you think you're making here. It is not physically possible to put as many seats behind the home plate as in bleachers 500 yards away. That's why many more people have to sit in the bleachers. It is certainly physically possible to build much more dense housing. The reason we don't build more dense housing is not that we can't but that there isn't the demand for it.

by Bertie on Aug 8, 2011 8:44 pm • linkreport

@Bertie, the paper you linked to is an opinion piece. This assertion in particular is in dispute if not outright refuted by scientific evidence, health science that is:

Third, sprawl’s negative quality of life impacts have been overstated. Effective vehicle pollution regulation has curbed emissions increases associated with increased driving. The growth of edge cities is associated with increases in most measures of quality of life.

by Tina on Aug 8, 2011 9:11 pm • linkreport

@Bertie, the paper you linked to is an opinion piece.

No, it's an academic research paper written by a Harvard professor of economics who specializes in urban economics, and published by the Harvard Institute of Economic Research.

This assertion in particular is in dispute if not outright refuted by scientific evidence, health science that is

Show us where that assertion is disputed in the scientific literature.

by Bertie on Aug 8, 2011 9:21 pm • linkreport

Most people seem to think that cheaper and newer is better.

How do you know people "think that cheaper and newer is better"? If so, they're probably wrong. What good is a cheap, new home if it's shoddily built and falls apart in 15 years? I don't think that is how people define better. Also, location matters. That is why, as you note, homes in the city are more expensive. Because of the land. And the land is more expensive because the location is better. You can build a real cheap home in the middle of nowhere, but you might have trouble selling it.

What evidence do you have that people are "ignoring" this cost?

"This paper describes how current residential lending practices encourage sprawl by ignoring the transportation cost savings provided by more urban locations."

Also, y'know, it's the fucking thesis of the report that this is all about

"This information gap on location efficiency, which is measured here as the cost of transportation associated with each place, leads to unexpected financial burdens and time constraints
for households, poor location decisions by developers, and missed and misplaced opportunities for municipalities."

That's how I know. Maybe you should take the time to read the report.

The evidence overwhelmingly shows that cars are generally much faster than mass transit. The average commute by transit takes more than twice as long as the average commute by automobile.

You're point was not about the average commute, but rather what "is quite obvious to people." I don't think it is and this argument kind of proves it. If it were obvious we'd all agree with you. But in much of DC, the car is not much faster or more convenient than mass transit. In fact, in many cities they have commuter races, and driving often loses.

For the vast majority of trips, biking and walking are even slower than transit.

That's absolute BS. Walking beats transit for all trips under 0.25 miles and up to a mile, which in the city, is a lot of trips. What about trips next door. And the bike is competitive up to 3-5 miles. so that is incorrect.

And it is moot. The point remains that you've ignored those trips and people often choose them over driving.

by David C on Aug 8, 2011 10:51 pm • linkreport

How do you know people "think that cheaper and newer is better"?

Because over the past 50 years there has been a massive migration from cities to suburbs and from old metropolitan areas in the northeast to new metropolitan areas in the south and west. The patterns of growth and migration demonstrate an overwhelming preference among Americans for suburban, car-oriented lifestyles.

If so, they're probably wrong.

You don't get to decide for other people what they think is better. If you prefer to live in a small, expensive apartment or condo in the city than in a new, spacious, affordable house in the suburbs, that's your choice. But most people clearly do not share your preference.

Also, location matters. That is why, as you note, homes in the city are more expensive. Because of the land. And the land is more expensive because the location is better.

No, as I said, housing in cities tends to be more expensive than housing in suburbs because land prices are higher in cities and construction costs are higher. Land prices are higher because density is higher, so there is more competition for each square foot of land. And construction costs are higher because it costs more to build vertically than horizontally. Mixed use development also tends to increase land costs because homebuyers are competing with commercial businesses who can afford to pay more.

You can build a real cheap home in the middle of nowhere, but you might have trouble selling it.

But I'm not talking about homes "in the middle of nowhere." I'm talking about housing in the suburbs. A growing majority of Americans live in suburbs.

by Bertie on Aug 8, 2011 11:28 pm • linkreport

Direct subsidies to automobile users are on the order of 1 cent per passenger-mile. Direct subsidies to mass transit users are on the order of 70 cents per p-m.

well, you haven't cited either of those numbers, but it doesn't matter. The point is moot.

Measuring per passenger mile isn't the right measure. Road building is cheaper because it is often done in the middle of nowhere, where, as you note, land is cheaper. Transit on the other hand is built almost exclusively in the middle of the freaking city. It has to be tunneled. You're comparing building a road in the middle of the Nevada dessert to building a subway line in Manhattan. As you put it, it's meaningless. It's not comparing like to like.

Furthermore, the purpose of transit is not to travel long distances – these are not pleasure travelers trying to get as far from home as possible, but rather commuters trying to get to wherever their jobs and schools are located. But the distance to this “somewhere” is not a variable to be held constant – it actually varies with population and job density, which is highly correlated with mode of transit. Places with train lines generally have and allow for denser development and thus less distance between your house and your workplace or school – the difference in average commute distance between urban and exurban areas could be as much as an order of magnitude.

If I ride the train 1 mile, I'm getting a 70 cent subsidy by your measure, but if you drive 35 miles each way, so are you. The correct measure is subsidy per trip.

And I showed that the indirect subsidy for driving was between $0.14 per mile and $0.43 per mile (20 mpg and 1.63 passengers, do the math). So it doesn't even need to be that much of a difference.

Let's note, though the indirect subsidy dwarfs the direct subsidy, you choose to compare the direct subsidy. You've chosen to compare only the tiniest slivers of the pie as though that tells the whole story. It doesn't. Because the larger part - and you've admitted that it's larger - highly favors transit.

I am aware that many buses use diesel (not sure if it's most. Would love to see that cited) but many do not. DC is 100% CNG I think. Seattle uses electricity as do other cities. Some places use fuel cells. And buses are not the only form of transit. The vast majority of streetcars, light rail and subways are electric. So the negative externalities of imported oil fall much more heavily on driving. And driving requires far more paving per person, so that too falls more heavily on driving. Safety, property damage, subsidized parking, - more heavily on driving. So I don't know what the indirect tax subsidy for transit is, but I'm confident that the ratio of direct to indirect is far lower than it is for driving.

And when coupled with the fact that drivers commute longer distances than transit users, I'm confident that the subsidy per trip is lower. And that's the measure that matters. So suck on that for a while.

by David C on Aug 8, 2011 11:29 pm • linkreport

The patterns of growth and migration demonstrate an overwhelming preference among Americans for suburban, car-oriented lifestyles.

But how do you know that they "think that cheaper and newer is better"? As you point out, you don't get to decide what they think is better. Without some actual polling data, you're making stuff up. Perhaps they just think it's cheaper, but not better.

You don't get to decide for other people what they think is better.

You've not proven that they think it's better. Only preferred. The two are not synonymous. I used to drive a Geo Metro, but I didn't think it was better than a BMW.

housing in cities tends to be more expensive than housing in suburbs because land prices are higher in cities and construction costs are higher.

Well then, why would anyone pay more for it? Do the laws of supply and demand somehow shut down in the city? They aren't cheaper. They aren't spacious. They aren't newere. If those are the things that make something better, why are people paying so much for them? They don't have to, do they? It must be because they're better right? That's why people usually pay more for something.

Your example of truffles and potatoes is crap. Obviously people think truffles are better, or else they'd just buy cheaper potatoes. Houses in the city are like truffles and houses in the exurbs are like potatoes.

The reason why houses in the city are more expensive is because that is what the market will bear. And the market will bear it, because people want it more. If people wanted to live in the exurbs more, they'd pay more for it and there would be bidding wars. That's how the market works.

Have you ever paid more for something that you didn't like as much as another choice? I haven't.

You're trying to prove that American prefer the suburbs over the city because they move there, while denying that they like the city more than the suburbs because they pay more. When what we have is an equilibrium - people vote with their feet and their dollars. Both groups are making the choice that they prefer. It's just that, as this report shows, many people who move to the suburbs are doing it without considering transportation costs which "leads to unexpected financial burdens and time constraints for household." So they're making uninformed choices.

by David C on Aug 8, 2011 11:54 pm • linkreport

"This paper describes how current residential lending practices encourage sprawl by ignoring the transportation cost savings provided by more urban locations."

You claimed that PEOPLE are ignoring the higher transportation costs of suburban living. Whether or not "current residential lending practises" encourage sprawl, that is not evidence that people are ignoring these costs. Where is your evidence for that assertion?

"This information gap on location efficiency, which is measured here as the cost of transportation associated with each place, leads to unexpected financial burdens and time constraints for households, poor location decisions by developers, and missed and misplaced opportunities for municipalities." That's how I know. Maybe you should take the time to read the report.

What "information gap in location efficiency?" The text you quote above appears nowhere in the paper you linked to. The fundamental problem with the report cited by David Alpert, which numerous commenters have pointed out, is that it compares costs but ignores benefits. It is meaningless to argue for urban lifestyles on the grounds that their combined housing and transportation costs are lower than those of suburban lifestyles without accounting for differences in the benefits of housing and transportation between cities and suburbs. Most obviously, the fact that suburban housing tends to be much cheaper per square foot, so that you can get a bigger home for a given price in the suburbs than you can in the city. And the fact that car travel is so much faster, more comfortable and more convenient than travel by mass transit or bike/walking.

You're point was not about the average commute, but rather what "is quite obvious to people."

I have no idea why you think it is not obvious to people that cars are faster than mass transit for the vast majority of trips. People are perfectly well aware that using buses and trains is generally slow. You have to get from the origin of your trip to the bus stop or train station, usually by walking. Then you have to wait for the bus or train to arrive. Then you have wait again while the bus or train stops at all the intermediate stops before it gets to the stop where you get off. Then you have to get from that stop to your final destination, again usually by walking. In the common event that your trip requires a transfer, you have to wait for a second bus or train to arrive. All of this generally takes a lot of time. That is why the average commute time by mass transit is more than double the average commute time by car. Transit beats cars on travel time only for a very small share of urban trips, typically rush-hour trips into or within the central areas of old dense cities like New York or DC. That is why even for commutes, in virtually every metropolitan area in the country, transit has less than a 10% share of commutes.

by Bertie on Aug 9, 2011 12:04 am • linkreport

Bertie, all I really need is explicit proof that the study you linked to accounts for general-fund municipal roads and capital costs of any source. Because it does not account for those costs.

The value of private automobiles is entirely in its ability to go point-to-point. If municipal roads and capital costs are not counted, then the argument has a hole in it.

by Neil Flanagan on Aug 9, 2011 12:08 am • linkreport

well, you haven't cited either of those numbers,

Yes I have. Transit subsidies are reported by the National Transit Database and road subsidies are reported in, for example, the paper by Mark Delucchi that I linked to.

Measuring per passenger mile isn't the right measure. Road building is cheaper because it is often done in the middle of nowhere, where, as you note, land is cheaper. Transit on the other hand is built almost exclusively in the middle of the freaking city. It has to be tunneled. You're comparing building a road in the middle of the Nevada dessert to building a subway line in Manhattan. As you put it, it's meaningless. It's not comparing like to like.

Another nonsensical argument. Mass transit pays nothing to use urban roads. There is no equivalent of the gas tax for transit users. The roads used by bus and urban light rail riders are funded by drivers and taxpayers. And taxpayers also pay most of the costs of building rail transit track and stations, purchasing bus and train vehicles, purchasing fuel, paying the wages of drivers and mechanics, and so on.

If I ride the train 1 mile, I'm getting a 70 cent subsidy by your measure, but if you drive 35 miles each way, so are you. The correct measure is subsidy per trip.

Huh? Why is it "correct" to subsidize by number of trips, but not by trip length? If it is "correct" to provide a fixed subsidy regardless of trip length, then transit fares should increase with the length of the trip. Once the fixed subsidy has been consumed, the transit user should pay the entire remaining cost of his trip. This would make long transit trips, such as suburb-to-city-center commutes, much more expensive for transit users, and provide a big incentive for current transit commuters to drive instead, and for employers to locate workplaces in the suburbs closer to where their employees live. That doesn't sound like what you want.

And I showed that the indirect subsidy for driving was between $0.14 per mile and $0.43 per mile

I have no idea where you think you showed any value for indirect subsidies for driving, let alone the numbers you state above. If you're alluding to the negative externalities of fuel and energy, then for about the third time, those negative externalities also apply to mass transit, and are additional subsidies to mass transit over and above the 70+ cents already provided in direct subsidies.

As I mentioned in an earlier comment, Delucchi has produced estimates of subsidies to mass transit users and automobile users that include both direct and indirect subsidies. In all cases, the total subsidies to mass transit users are VASTLY higher than the total subsidies to automobile users, both in cents per passenger-mile and as a share of total costs. There is simply no way you can seriously claim that subsidies incentivize driving over using mass transit.

by Bertie on Aug 9, 2011 12:51 am • linkreport

But how do you know that they "think that cheaper and newer is better"?

For about the third time, from their choices as voters and consumers. Economists call it "revealed preference." If people thought urban lifestyles were better than subsurban ones, they would choose urban lifestyles. Overwhelmingly, they make the opposite choice.

You've not proven that they think it's better. Only preferred. The two are not synonymous

You now seem to be reduced to playing silly word games. I'm not sure why you think someone would prefer the choice they considered worse, but it is completely irrelevant to the point. For reasons that seem rather obvious but that you seem incapable of grasping -- cheaper, better housing; the speed and convenience of car travel -- Americans are overwhelmingly choosing suburban lifestyles over urban ones. That doesn't seem likely to change.

Well then, why would anyone pay more for it?

Because some people think that other aspects of urban lifestyles compensate for the expensive, cramped housing and other disadvantages. These people are mainly young, single, childless adults. Also, recent immigrants. It's not that there's no market at all for urbanism. The point is that it's a small, niche market. As a share of the total market for housing and transportation, it's shrinking.

Your example of truffles and potatoes is crap. Obviously people think truffles are better, or else they'd just buy cheaper potatoes. Houses in the city are like truffles and houses in the exurbs are like potatoes.

You have completely missed the point. Truffles are much more expensive than potatoes. That doesn't mean there's more demand for truffles than potatoes. It's a result of the fact that truffles cost much more than potatoes to supply. Similarly, dense urban housing is more expensive than suburban housing because dense urban housing costs more supply.

The reason why houses in the city are more expensive is because that is what the market will bear.

No, the reason why housing in the city is more expensive is because it costs more to supply. Land prices are higher and construction costs are higher. The high cost of supply means that builders must charge higher prices to cover their costs, which in turn limits the size of the market.

You're trying to prove that American prefer the suburbs over the city because they move there, while denying that they like the city more than the suburbs because they pay more.

You're still totally confused. Most Americans DON'T "pay more" to live in the city. It's not worth it to them. They choose to live in the suburbs instead. Only a minority of Americans are willing to pay the higher price of living in the city. It's a niche market.

When what we have is an equilibrium - people vote with their feet and their dollars. Both groups are making the choice that they prefer. It's just that, as this report shows, many people who move to the suburbs are doing it without considering transportation costs

You have produced no evidence whatsoever that "many people who move to the suburbs are doing it without considering transportation costs." This is just pure invention on your part. You might just as well say that "many people who move to the city are doing it without considering housing costs and the inconvenience of mass transit."

by Bertie on Aug 9, 2011 1:31 am • linkreport

Bertie, all I really need is explicit proof that the study you linked to accounts for general-fund municipal roads and capital costs of any source. Because it does not account for those costs.

Yes it does. Delucchi's analysis includes all spending on roads by "federal, state and local governments." That includes spending that comes out of the general fund. I'm not sure what spending you think Deluccchi is excluding, or what basis you think you have for that claim.

by Bertie on Aug 9, 2011 1:50 am • linkreport

I forget who said this (Mark Twain? Vin Scully?), but it seems appropriate to point out:

"Statistics are used much like a drunk uses a lamppost: for support, not illumination."

Yes, Bertie, I'm looking at you.

by dcd on Aug 9, 2011 7:52 am • linkreport

No, the reason why housing in the city is more expensive is because it costs more to supply. Land prices are higher and construction costs are higher. The high cost of supply means that builders must charge higher prices to cover their costs, which in turn limits the size of the market.

Because, of course, the only kind of housing, in cities or suburbs, is brand-new housing.

by MIriam on Aug 9, 2011 9:05 am • linkreport

Yes I have. Transit subsidies are reported by the National Transit Database and road subsidies are reported in, for example, the paper by Mark Delucchi that I linked to.

You linked to those, but didn't cite them as the source for your subsidy per passenger mile numbers.

Mass transit pays nothing to use urban roads. There is no equivalent of the gas tax for transit users. The roads used by bus and urban light rail riders are funded by drivers and taxpayers. And taxpayers also pay most of the costs of building rail transit track and stations, purchasing bus and train vehicles, purchasing fuel, paying the wages of drivers and mechanics, and so on.

Nonsense. Many bus systems make payments to localities for roads. Many, if not most, transit systems are not exempt from gas taxes or taxes on electricity. As for all these other subsidies, aren't you including them in your number? If not, why not tell the whole story? If so, you're double dipping. No fair.

Why is it "correct" to subsidize by number of trips, but not by trip length?

Because the goal of going to work is not to travel a lot of miles, but to get to work.

Say I take two friends to dinner, and I tell them I will subsidize steak at $5 per steak and beer at $1 per steak. If person one orders a steak and no beers and the other orders 6 beers and no steak, what do I care that person two got a lower subsidy per unit. He got a higher overall subsidy.

Drivers are like the beer drinker. They may get a lower subsidy per unit, but they use a lot more units. So the subsidy per trip is higher.

But it goes farther than that. That transit user is more likely to do more trips on foot, which gets an even lower subsidy. So we really need to look at subsidy per person per year. How much do we subsidize each lifestyle choice. I don't care that you only get a 1 cent per passenger mile subsidy if you then go out and drive 15,000 compared to my 500 miles of transit use.

transit fares should increase with the length of the trip.

They often do.

I have no idea where you think you showed any value for indirect subsidies for driving, let alone the numbers you state above.

You have to do some math. I cited the true cost of gasoline per gallon in 1998. Less the cost of gasoline in 1998 ($1 per gallon). Then divide that by average fuel milage and passengers per mile. Sorry. I thought you could handle that. I won't overestimate you again.

If you're alluding to the negative externalities of fuel and energy, then for about the third time, those negative externalities also apply to mass transit, and are additional subsidies to mass transit over and above the 70+ cents already provided in direct subsidies.

Yes. I get that. For the third time, how much is that per passenger mile? It isn't the same as for driving for the reasons I listed above.

In all cases, the total subsidies to mass transit users are VASTLY higher than the total subsidies to automobile users, both in cents per passenger-mile and as a share of total costs.

And again, what did he come up with?

You claimed that PEOPLE are ignoring the higher transportation costs of suburban living. Whether or not "current residential lending practises" encourage sprawl, that is not evidence that people are ignoring these costs. Where is your evidence for that assertion?

I didn't realize that current lending practices were carried out by robots. Even if you think current leanding practices don't influence people's decisions, the report this post is all about makes it clear that "leads to unexpected financial burdens and time constraints for households". Are households not people either?

The text you quote above appears nowhere in the paper you linked to.

It's in the paper David cited.

And the fact that car travel is so much faster, more comfortable and more convenient than travel by mass transit or bike/walking.

Not a fact. It's your bias.

If people thought urban lifestyles were better than subsurban ones, they would choose urban lifestyles. Overwhelmingly, they make the opposite choice.

Oh. I see. What people choose most is "better". So the fact that more people worldwide choose to live in slums than choose to live in suburbs makes slums better than suburbs. Interesting. I guess a Ford Taurus is better than a BMW. And Dells are better than Apples. And McDonald's hamburgers are better than lobster.

That's not how I define better, and I don't think anyone else does either. That's what I meant by, it depends on how you define better.

I'm not sure why you think someone would prefer the choice they considered worse

Becasue they can't afford they one they think is better. Or they don't think they can afford it, because they don't consider transportation costs.

For reasons that seem rather obvious but that you seem incapable of grasping -- cheaper, better housing

Just because you repeat something over and over again, it doesn't make it true. You couple cheaper here with better. Even though better for you means "what people mostly choose" and people mostly choose the suburbs because it's cheaper. So that's completely fucking redundant.

the reason why housing in the city is more expensive is because it costs more to supply.

Markets raise prices to the level that customers will pay. That's why housing is more expensive. Or do you think houses in the suburbs are cheaper, despite the fact that would pay as much for them as they'd pay for a house in the city?

Land prices are higher

because of higher demand.

Most Americans DON'T "pay more" to live in the city. It's not worth it to them.

Because they can't afford it. Just like a BMW.

You have produced no evidence whatsoever that "many people who move to the suburbs are doing it without considering transportation costs."

I cited two reports. You aren't even paying attention anymore. You're clearly very confused. Perhaps you should take a nap and stop reading the drivel that idiot Randall O'Toole shills out.

by David C on Aug 9, 2011 10:02 am • linkreport

This is a fascinating debate and I hope you all keep going, but I want to remind everyone please not to use insults. I strongly disagree with O'Toole and think his methods are suspect, but we should not call him or anyone an "idiot." Also, the more this discussion can stick to pointing out information rather than using phrases like "your bias," "you're playing word games," "your argument is crap," "you're confused," and so forth.

by David Alpert on Aug 9, 2011 10:13 am • linkreport

You have completely missed the point. Truffles are much more expensive than potatoes. That doesn't mean there's more demand for truffles than potatoes. It's a result of the fact that truffles cost much more than potatoes to supply. Similarly, dense urban housing is more expensive than suburban housing because dense urban housing costs more supply.

Your metaphor is inapt. Potatoes and truffles are not substitutable. As David C pointed out, a more appropriate metaphor is automobiles. There's both a utility and a personal preference aspect operating here.

If you've got five kids, you're not going to get a two-seat sports-car. You're going to get a minivan or an SUV. That's the utility component. But the more desirable the vehicle within the appropriate class, the more you'll pay for it.

The car metaphor is even more apt given that the narrative of the young man forced to give up his beloved sports-car in favor of a mini-van is a common cliche. It's the same with the couple who are heartbroken to move out of the city because they have no decent public school options. A sizable percentage of those folks who "choose" the suburbs do so because they feel they have no other choice. The lack of suitable schools chooses for them. That certainly has nothing to do with personal preference.

Not only does your "potatoes/truffles" metaphor fail the basic test of comparability, but it's obviously a tissue-thin attempt at prejudicing the listener. After all what could be more all-American than the potato. Hell, I ain't even seen a truffle before. Don't need any of that fancy-Nancy stuff!

It's the equivalent of asking, "Which do Americans prefer: Bread or Extra Virgin Olive Oil?" or "Which do Americans like better, clean American spring water or an expensive French wine?"

by oboe on Aug 9, 2011 10:26 am • linkreport

O'Toole isn't even something anyone CAN cite - he doesn't give any sort of background on his methods or how he comes up with any of the numbers he uses.

Bertie, your entire argument just consists of a bunch of circular logic. You say "people choose suburban living because it's better," and then when challenged on this come right back with "it's clearly better because more people are choosing it!"

If you're going to try to argue that transit is way less efficient than driving, and you actually want to compare apples to apples (transit vs commuter driving), then you're mistaken. Transit gets 41 passenger miles per gallon equivalent, and I'm going to guess based on CAFE that people driving to work alone in their cars are not doing that well. And again, the overall transit number (41pmpg) is worse than the transit commuting number, since it includes off-peak, mobility service, etc. It also doesn't include the direct benefit from transit like the gas saved because people are using transit instead of driving (340 million gallons of gas) or the benefits of trip distance reduction due to land use changes attributable to transit (3.4 billion gallons of gas).

(APTA Fact Book for these numbers: http://www.apta.com/resources/statistics/Pages/transitstats.aspx)

Not sure what your entire point is with this ongoing discussion - do you just feel personally insulted because you chose to live in the suburbs? That's not anyone's intent here.

by MLD on Aug 9, 2011 10:43 am • linkreport

Becasue they can't afford they one they think is better. Or they don't think they can afford it, because they don't consider transportation costs.

c.f. DCPS. Nothing to do with housing type preference. Everything to do with cheap schooling. Fix the schools, and even with the massive subsidies we provide to suburbia, dense walkable communities will rate even greater price premiums than in comparison to the exurbs.

by oboe on Aug 9, 2011 10:49 am • linkreport

Can we stop with this growth rate non-sense? Of course Loudoun County has a higher growth rate than Arlington. There are thousands and probably millions of open undeveloped space in Loudoun County, where there is so much less open develop able space in Arlington or in DC for that matter.

So here is a BS stat from the 2010 census: http://quickfacts.census.gov/qfd/states/51/51107.html

Loudon County population: 312,311
Washington, DC population: 601,723

So because more people live in the urban city core of the region it is obvious that it is a better place to live than Loudoun County

by Ryan on Aug 9, 2011 10:52 am • linkreport

@Bertie, the Glaeser paper you linked to is a discussion paper. It says so right there in the title and in the first sentence presents itself as an "essay". Its not a research paper. It does not present any original research. It does not begin with a research question. The authors state their purpose as not to investigate a research question or a hypothesis but rather, In this essay, we review the economic literature on sprawl and urban growth, and make
four points.

This is not research. Its rhetoric or opinion. Furthermore not one referrence in the bibliography is a health research paper reporting results on auto-emmissions and health, the built environment and health, and how policies and zoning contribute to those outcomes.

Any essay that claims to make diffinitive "points" about quality of life, urban sprawl and policies without mentioning human health consequences or impacts is incomplete either purposefully, as if the author had an opinion to bolster by ignoring those data, or by incompentence.

There are a plethora of publically available manuscripts reporting results from well designed hypothesis driven original research studies informing us of the negative consequences to human health associated with car-dependent built environments and life-styles. I don't have time to provide you with a bibliography right now. You could start by going to pubmed and using key words "built-environment, heart diesease, air pollution, obesity, diabetes, asthma, preventable death".

by Tina on Aug 9, 2011 11:00 am • linkreport

And, I forgot to mention, the economic burdens both to individuals/households and to communities that result from the negative health outcomes associated with increased auto traffic emissions, unwalkable built environments, general car dependency and inadequate physical activity. Again, Glaeser et. al blithely ignore the health economics of urban sprawl in their essay.

by Tina on Aug 9, 2011 11:57 am • linkreport

I find this whole discussion a waste of time.

I don't have loads of statistics or reports to back my thoughts---but I do have common sense.

Anyone who looks back at DC area real estate market trends even casually, for the past century can see that starting after WWII, many DC residents started to move to the suburbs of VA and MD. This trend continued up through most of the 50s, the 60s (the late 60s riots didn't help), and 70s. Starting in the mid-late 80s, the trend began to reverse (slowly) so people started wanting to move back into the city. It's clear to see..just watch the development of neighborhoods like Dupont Circle, Logan Circle, etc..into the 90s.

The first decade of the 21st century has brought even further growth to DC, with more neighborhoods like Penn Quarter (previously not residential), Columbia Heights, Petworth, Adams Morgan, etc, etc...increasing in population.

Perhaps I am wrong--but just looking around, it seems pretty clear that while at some point Bertie's assumptions might have been true (I'd say from 50s-late 80s)...I don't think they are true anymore.

I think the average DC area resident is willing to forgo some living space, for more convenience, and most importantly, a faster/easier commute to work.

And no, I don't have reports/stats/figures to back it up. That's my assessment from observing the DC area, being born here and having lived here most my life. So sue me.

by LuvDusty on Aug 9, 2011 1:18 pm • linkreport

Many bus systems make payments to localities for roads. Many, if not most, transit systems are not exempt from gas taxes or taxes on electricity.

This claim is simply false. Gas taxes are levied on retail sales of gasoline. Transit agencies do not buy gasoline from retail outlets. They buy it from suppliers at wholesale prices. Transit users pay nothing to use roads. Their use of roads is entirely subsidized by drivers and taxpayers.

Because the goal of going to work is not to travel a lot of miles, but to get to work. Say I take two friends to dinner, and I tell them I will subsidize steak at $5 per steak and beer at $1 per steak. If person one orders a steak and no beers and the other orders 6 beers and no steak, what do I care that person two got a lower subsidy per unit. He got a higher overall subsidy.

Huh? How does "the goal of going to work is not to travel a lot of miles, but to get to work" mean that it is "correct" to subsidize transportation in units of trips rather than passenger-miles? What does "the goal of going to work" even have to do with trips that have nothing to do with work?

They may get a lower subsidy per unit, but they use a lot more units.

Again, so what? If you think people who "use more units" should not receive more in subsidies than people who use fewer units then you should be advocating a fixed subsidy per person regardless of the number of "units" they consume, whether the units are passenger-miles, trips or anything else. Is this now your position?

They often do.

They usually don't. And even when they do, the increase in fares is not sufficient to cover the increase in costs. In general, transit fares are fixed regardless of trip length. A person who takes a 10-mile bus trip receives much more in subsidies than a person who takes a 1-mile bus trip. If it is "correct" to provide higher subsidies to people who consume more miles of travel by transit, why isn't it "correct" to provide higher subsidies to people who consume more miles of travel by automobile?

Yes. I get that. For the third time, how much is that per passenger mile? It isn't the same as for driving for the reasons I listed above.

You haven't produced any data or analysis that compares indirect subsidies for automobile users versus transit users. Therefore, you have no basis for claiming that indirect subsidies are any higher for auto users than for transit users, let alone that they are so much higher for auto users as to offset the enormously higher direct subsidies provided to transit users. As I have already noted, Delucchi has estimated both direct and indirect subsidies for auto users and transit users and concluded that total subsidies are overwhelmingly higher for transit users, both in cents per passenger-mile and as a share of total costs.

by Bertie on Aug 9, 2011 2:17 pm • linkreport

Transit users pay nothing to use roads. Their use of roads is entirely subsidized by drivers and taxpayers.

Can anyone spot the fallacy here?

by oboe on Aug 9, 2011 2:20 pm • linkreport

Can we stop with this growth rate non-sense? Of course Loudoun County has a higher growth rate than Arlington. There are thousands and probably millions of open undeveloped space in Loudoun County, where there is so much less open develop able space in Arlington or in DC for that matter.

The mere fact that one county has more "open undeveloped space" than another does not mean we should expect it to grow at a higher rate. There are numerous counties with more open undeveloped space than either Loudon or Arlington but that have grown more slowly than both. The relevance of the fact that Loudon, and the other outer suburbs of DC, have grown much faster than Arlington and DC itself is that it demonstrates the overwhelming preference for low-density, car-oriented lifestyles over high-density, transit-oriented lifestyles. The same pattern has occurred in almost every other metropolitan area in the country. Contary to the "back to the city" narrative promoted by proponents on urbanism, and despite huge government spending on new transit infrastructure in urban cores, the most recent Census data clearly shows that the longstanding trend of suburbanization and sprawl has continued during the most recent decade, in the DC metro area as well as throughout the country as a whole.

by Bertie on Aug 9, 2011 2:33 pm • linkreport

C'mon oboe, haven't you seen the special box you check on your tax forms that says transit rider and exempts you from taxes?

But beyond that, Bertie's assertion that transit agencies don't pay fuel taxes is unfounded. All I could easily find was this CTAA paper about it, but it would appear plenty of agencies are not exempted from gas taxes. I would assume that most large transit agencies apply for the exemption though.

A person who takes a 10-mile bus trip receives much more in subsidies than a person who takes a 1-mile bus trip. If it is "correct" to provide higher subsidies to people who consume more miles of travel by transit, why isn't it "correct" to provide higher subsidies to people who consume more miles of travel by automobile?

This really isn't the right way to look at it. I could do the math to show you that every transit trip regardless of length receives the same subsidy (just divide subsidy $ by number of trips provided). You might say this is wrong because all trips are not the same, but I can also say that all passenger miles (your measure) are not the same. One passenger mile in a dense urban area is not the same as one passenger mile in the middle of nowhere, at least in terms of the cost of providing that service.

Delucchi has estimated both direct and indirect subsidies for auto users and transit users and concluded that total subsidies are overwhelmingly higher for transit users, both in cents per passenger-mile and as a share of total costs.

The paper you're talking about does not cover all the negative externalities people are talking about. It tries to quantify all the government spending on roads and look at that in comparison to user fees. It does not quantify the environmental impact of car travel or anything like that. It clearly says this in the abstract of the paper.

by MLD on Aug 9, 2011 2:45 pm • linkreport

Potatoes and truffles are not substitutable.

Whether they are "substitutable" is irrelevant. The point is that you cannot infer relative demand from price comparisons alone. Just as the fact that the price of truffles is higher than the price potatoes does not mean there is more demand for truffles than potatoes, the fact that the price of dense urban housing is higher than the price of suburban housing does not mean there is more demand for dense urban housing than suburban housing. The reason dense urban housing is more expensive is because it costs more to supply. That means builders have to charge higher prices to cover their costs. I don't know why you can't understand this basic point. If apples cost $1 each to produce and oranges cost $2, farmers must charge at least $2 per orange to cover their costs, but only $1 per apple. This difference in pricing has nothing to do with the relative level of demand for apples vs. oranges. It's dictated by the costs of production.

by Bertie on Aug 9, 2011 2:52 pm • linkreport

@Bertie: "The relevance of the fact that Loudon, and the other outer suburbs of DC, have grown much faster than Arlington and DC itself is that it demonstrates the overwhelming preference for low-density, car-oriented lifestyles over high-density, transit-oriented lifestyles."

No, it demonstrates a preference for large homes that they can't afford in the city. The car-oriented lifestyle isn't something they seek out; it's something they put up with because something else is more important. Correlation =/= causation.

At some point, though, the car oriented lifestyle becomes so soul-sucking that avoiding it becomes more important than the shitty constriction, 6 bedroom McMansion in [insert outer suburb/exurb]. I think the crux of this debate is that you believe that day will never come. Others, myself included (hell, I've opted out of it already), disagree.

I would like to ask, though, what are you proposing? It seems like you're advocating shutting down Metrorail (and Metrobus?) over a 20 year period or so, and investing the savings in roads. Land use planning should be revised to favor sprawl (not that it doesn't already in the outer burbs.) Is that right?

by dcd on Aug 9, 2011 2:56 pm • linkreport

@MLD:

The paper you're talking about does not cover all the negative externalities people are talking about.

Funny, I was just about to point that out. In fact, it's so relevant to Delucchi's numbers that it appears in the abstract.

by oboe on Aug 9, 2011 2:57 pm • linkreport

@Bertie:

The reason dense urban housing is more expensive is because it costs more to supply.

Look, this is getting silly. Your logic is perfectly circular. Your argument is that urban housing is more expensive because it costs more.

My argument is that it's more expensive because the land is more desirable. In other words it's location, location, location. The desirability of the land is closely tied to the value of the housing stock. Don't believe me? Build a 200 unit condo out in rural WVA and try charging $600 per sq ft. :)

by oboe on Aug 9, 2011 3:04 pm • linkreport

This really isn't the right way to look at it. I could do the math to show you that every transit trip regardless of length receives the same subsidy (just divide subsidy $ by number of trips provided).

No you couldn't. Long trips cost more to supply than short trips. Fuel costs are higher, labor costs are higher, equipment costs are higher. If transit users pay the same fare regardless of the length of the trip, then they receive a higher subsidy for longer trips. If it is "correct" for transit users to receive a higher subsidy for longer trips, why isn't it "correct" for auto users to receive a higher subsidy for longer trips?

You might say this is wrong because all trips are not the same, but I can also say that all passenger miles (your measure) are not the same. One passenger mile in a dense urban area is not the same as one passenger mile in the middle of nowhere, at least in terms of the cost of providing that service.

This is completely irrelevant. Even if we pretend that road costs apply only to roads in urban areas, transit users are still subsidized at a vastly higher rate per passenger-mile than auto users. Again, the disparity is on the order to 70-to-1. There is simply no way you can seriously claim that subsidies incentivize driving over using mass transit.

The paper you're talking about does not cover all the negative externalities people are talking about. It tries to quantify all the government spending on roads and look at that in comparison to user fees. It does not quantify the environmental impact of car travel or anything like that. It clearly says this in the abstract of the paper.

I'm not talking about the paper I linked to regarding road subsidies. I'm talking about other research by Delucchi in which he addresses negative externalities such as pollution.

by Bertie on Aug 9, 2011 3:08 pm • linkreport

Look, this is getting silly. Your logic is perfectly circular. Your argument is that urban housing is more expensive because it costs more.

You still don't seem to understand. A higher cost of supply means a higher market price. There's nothing "circular" about it. It's basic economics. How do you think builders can cover their costs, let alone make a profit, unless they charge at least as much for their houses as it costs to build them?

My argument is that it's more expensive because the land is more desirable.

No, the land is more expensive because there are more people competing for it. Higher density means more people per unit of land. Higher demand per unit of supply means a higher unit price. And in addition to higher land costs, dense urban housing has higher construction costs. It costs more to build vertically than horizontally.

by Bertie on Aug 9, 2011 3:17 pm • linkreport

@Bertie,

My argument is that it's more expensive because the land is more desirable.

No, the land is more expensive because there are more people competing for it.

--------------------------------------------

If more people are competing for it, how the **** is not more desirable?

by Ryan on Aug 9, 2011 3:26 pm • linkreport

@Bertie,

So, in other words, a two-unit condo is more expensive to build than a couple of single-family homes, irregardless of land costs. Got it.

That seems quite wrong to me.

In fact, it always seemed to me that suburban housing was cheaper because the land was, in effect, worthless.

by oboe on Aug 9, 2011 3:27 pm • linkreport

Hahaha I'm done. Good luck friends!

by MLD on Aug 9, 2011 3:30 pm • linkreport

@MLD,

Agreed, I think we may have finally hit groundwater.

by oboe on Aug 9, 2011 3:32 pm • linkreport

No, it demonstrates a preference for large homes that they can't afford in the city.

Yes, bigger housing is one aspect of suburban lifestyles that makes them so much more popular than urban lifestyles.

The car-oriented lifestyle isn't something they seek out; it's something they put up with because something else is more important. Correlation =/= causation.

Of course they seek it out. But the fact that most people prefer a suburban lifestyle doesn't mean they think it's better than urban living in every single respect. They prefer it because they think it's better overall. The advantages of suburban life that people tend to cite most often are more affordable housing and the greater mobility of car travel. Other advantages of the suburbs include more privacy, less noise, a cleaner environment, less stress, better schools, and generally a higher quality of life.

I would like to ask, though, what are you proposing?

A reduction in the enormous subsidies provided to mass transit. But even with those subsidies, proponents of urbanism and mass transit are fighting a losing battle. Americans are continuing to suburbanize and sprawl despite the huge government incentives to bring them back to the city.

by Bertie on Aug 9, 2011 3:34 pm • linkreport

This claim is simply false. Gas taxes are levied on retail sales of gasoline. Transit agencies do not buy gasoline from retail outlets.

Or it is simply true. According to this FHWA site, transit use is taxed, at least in part, in: Alaska, Arizona, Colorado, Florida, Hawaii, Kansas, Louisiana, Maryland, Minnesota, Missouri, Montana, New Hampshire, New Mexico, Ohio, South Carolina, Texas and Virginia.

But other than that, you're right.

Look if you can't even get simple facts right, why should we expect you to get complex analysis right?

How does "the goal of going to work is not to travel a lot of miles, but to get to work" mean that it is "correct" to subsidize transportation in units of trips rather than passenger-miles?

The goal of transportation is transporting things and people from point A to point B. If someone chooses to live in West Virginia and shop in Delaware or work in DC, they use a much larger subsidy than someone who lives, works and shops in Alexandria, VA. So, one's choice to live far from the places they go means they use more subsidy. One's choice to live close means they use less.

If one person chooses to live 35 miles from work in an area that requires both driving to work and another 30 miles a day of driving for errands, they're using a larger subsidy then someone who can walk 1/4 of a mile to work and walk to do all their errands, excpet for once a week when they ride transit to church.

So even though the two might make the same number of trips, and the second person's transit trip is more heavily subsidized, the first person's total transportation needs are more subsidized.

What does "the goal of going to work" even have to do with trips that have nothing to do with work?

Work is one kind of trip. I was using it as an example.

Again, so what?

That seems pretty obvious. Using many units of a subsidized resource is taking a lot of tax dollars.

If you think people who "use more units" should not receive more in subsidies than people who use fewer units then you should be advocating a fixed subsidy per person regardless of the number of "units" they consume, whether the units are passenger-miles, trips or anything else. Is this now your position?

No. My position is that we should build the most efficient transportation system we can. Having people live near the things they go to, and then walking or biking to them, is the best way to do that. Driving - and the intensive amount of land that driving requires - is not conducive to that kind of efficiency, but transit is. We should not subsidize a lifestyle that requires driving over long distances because that is inefficient.

Better yet, we should not subsidize anything, but should charge the total negative externalities associated with each behavior. That means a gas tax that pays for all the pollution that driving (and transit) create. 100% tolled roads. Congestion pricing. A carbon tax. A surtax to pay for military expenditures to protect oil flow. Stormwater runoff fees for parking. Etc... If you did that, driving would be much more expensive, transit would be somewhat more expensive and walking and biking would remain pretty much the same price as now (free). That would make living in walkable and bikable communities much more competitive. And those communities are the same ones where transit makes sense. So transit would be just fine. Because the added cost of using transit occasionally would be dwarfed by the savings from not driving 15000 miles a year.

That's my position.

If it is "correct" to provide higher subsidies to people who consume more miles of travel by transit, why isn't it "correct" to provide higher subsidies to people who consume more miles of travel by automobile?

Neither is correct. I think we should charge transit per mile. For buses, that is technically difficult and for most rail systems as well, but I think it would be better. If you can figure out how to make it work, I'd support it.

You haven't produced any data or analysis that compares indirect subsidies for automobile users versus transit users.

True. Neither have you.

Therefore, you have no basis for claiming that indirect subsidies are any higher for auto users than for transit users, let alone that they are so much higher for auto users as to offset the enormously higher direct subsidies provided to transit users.

I do have a basis. It is called logic.

1. The indirect subsidy is much larger than direct subsidy.
2. The indirect subsidy is based largely on gasoline, but also on pavement, safety and health
3. Transit uses much less gasoline/diesel per mile than driving and much less pavement. It is safer and healthier (less stress, more walking).

If those three statements are true than the indirect subsidy for transit is smaller per mile than driving. Do I know how large the indirect subsidy is? No. But I've given a pretty good reason to believe that the total subsidy for transit may be less than the total subsidy for driving - more so on a per trip or per person basis. Especially considering (1).

Comparing the direct subsidy from one to the other is like trying to figure out who will win a football game by comparing the quality of their punters. It is a very small part of the picture. Nonetheless, you continue to try to makes some sort of deduction based on this trivial fact.

As I have already noted, Delucchi has estimated both direct and indirect subsidies for auto users and transit users and concluded that total subsidies are overwhelmingly higher for transit users

You have noted it. You have not linked to this mythical Delucchi study. The study you linked to has this disclaimer "Note that in this accounting we include only government expenditures; we do not include any ‘‘external’’ costs of motor-vehicle use." [In that same study, he puts the subsidy to driving at 20-70 cents per gallon which is 1-3 cents per mile, higher than what you stated, btw]

So, considering how wrong you were about transit systems being exempt, you'll excuse me if I don't take your word on this one right?

Lay in on the table. What is the total subsidy for both? Without it, we aren't talking facts.

by David C on Aug 9, 2011 3:36 pm • linkreport

"more desireable" equates to more people wanting it, or, "more people competing for it". Its the same thing.

Also -what dcd said about car dependency and the exurbs/suburbs. People are not necessarily choosing car dependency. That claim is spurious. For an example see the ridership of the expanded VRE, that outpaced projections. Why would so many people who live in exurbs ride the train in unexpected numbers? A: because it became available to them.

by Tina on Aug 9, 2011 3:37 pm • linkreport

The advantages of suburban life that people tend to cite most often are more affordable housing and the greater mobility of car travel.

You have produced no evidence whatsoever that people cite these things. This is just pure invention on your part.

by David C on Aug 9, 2011 3:47 pm • linkreport

It strikes me that everything we've said about Washington could be said about Paris. Housing prices per square foot are ridiculously high compared to the cités. And the suburbs have been growing at a startling rate compared to the city. As this graph shows, the so-called "City of Lights" has completely stagnated over the last century:

From this we can conclude that the French are overwhelmingly choosing to make their home here...

...rather than here...

by oboe on Aug 9, 2011 3:49 pm • linkreport

@Bertie: "Other advantages of the suburbs include more privacy, less noise, a cleaner environment, less stress, better schools, and generally a higher quality of life."

Cleaner environment - Do you really believe that having everyone depend on a car to get everywhere actually is BETTER for the environment? Oy.

Less stress - I don't think most people think, "My doctor said I need to reduce my stress - what to do, what to do . . . I know! I'll add a 60 minute commute, and begin and end of my day in virtual gridlock. Ten hours a week in heavy traffic is just the thing to bring my blood pressure down!" To each his own, I guess.

Quality of life - that's pretty subjective, isn't it? I firmly believe my quality of life if leaps and bounds better in the city than it would be Loudoun, et al.

by dcd on Aug 9, 2011 3:50 pm • linkreport

David C:

You just advocated a per-trip subsidy rather than a per-passenger-mile subsidy. That means you oppose the current subsidy policy for transit users, which provides a higher subsidy for longer trips. A fixed per-trip subsidy means that transit users would pay the entire costs of their trip in excess of the fixed subsidy. This means you support a dramatic increase in transit fares for long transit trips, such as commuter rail trips from suburbs to city centers. A dramatic increase in commuter rail fares would encourage current transit commuters to switch to driving instead.

I can find no clear argument for your per-trip subsidy position in your rambling comments. Why do you think transportation should be subsidized at all? Why do you think it should be subsidized in units of trips rather than passenger miles? Nowhere do you explain this. But then later, you write "Better yet, we should not subsidize anything, but should charge the total negative externalities associated with each behavior." If "we should not subsidize anything" then all transportation users should pay the full costs of their travel, both direct costs and indirect costs (including, but not limited to, negative externalities). But that position contradicts your previous advocacy of a per-trip subsidy. And then later still you write "I think we should charge transit per mile." But you don't say whether you think this "charge" should be subsidized or paid in full by the user. In short, your position on subsidies is completely incoherent.

by Bertie on Aug 9, 2011 4:40 pm • linkreport

You just advocated a per-trip subsidy rather than a per-passenger-mile subsidy.

I did no such thing. I said when calculating the subsidy for comparison purposes you should compare the per trip subsidy, not the per mile subsidy.

A dramatic increase in commuter rail fares would encourage current transit commuters to switch to driving instead.

Or it would encourage them to move closer to work. But I'm not advocating that. I'm advocating getting rid of all subsidies AND pricing all negative externalities.

Why do you think transportation should be subsidized at all?

I don't.

Why do you think it should be subsidized in units of trips rather than passenger miles?

I don't. But I think we should compare trips rather than miles and I've already made the best attempt to explain why above.

But that position contradicts your previous advocacy of a per-trip subsidy.

I never advocated such a subsidy. Show me where I did that.

But you don't say whether you think this "charge" should be subsidized or paid in full by the user.

I do. In fact you just quoted me saying "we should not subsidize anything, but should charge the total negative externalities associated with each behavior." So that's where I say it.

It takes two to communicate, and unfortunately you aren't doing your part.

I'll. Many of us called you out on your glaring mistake about transit fuel being exempt from taxes. And your inability to cite this paper that proves your point about indirect subsidies being greater for transit. You've failed to address these errors. In short, you're not arguing honestly. So I see no point in wasting my time with someone who, in addition to being able to comprehend simple ideas, is either willfully wrong or completely unwilling to find out the validity of a fact before asserting it. So, I'm done.

There are none so blind...

by David C on Aug 9, 2011 4:51 pm • linkreport

@Bertie -you stopped reading David Cs comment too soon. You seemed to have missed this:

...we should not subsidize anything, but should charge the total negative externalities associated with each behavior. That means a gas tax that pays for all the pollution that driving (and transit) create. 100% tolled roads. Congestion pricing. A carbon tax. A surtax to pay for military expenditures to protect oil flow. Stormwater runoff fees for parking. Etc... If you did that, driving would be much more expensive, transit would be somewhat more expensive and walking and biking would remain pretty much the same price as now (free).

by Tina on Aug 9, 2011 4:57 pm • linkreport

oboe

It strikes me that everything we've said about Washington could be said about Paris. Housing prices per square foot are ridiculously high compared to the cités. And the suburbs have been growing at a startling rate compared to the city. As this graph shows, the so-called "City of Lights" has completely stagnated over the last century:

Correct. The Paris metropolitan area has undergone massive suburbanization and sprawl over the past 50 years, just like most other large cities in the U.S. and Europe.

From this we can conclude that the French are overwhelmingly choosing to make their home here...

No, we can't conclude that. The picture you posted appears to be one of the many high-rise public housing projects that have been constructed in the poverty-stricken, crime-ridden inner suburbs of Paris to house the large numbers of poor immigrants that have been arriving from France's former colonies. I'm not sure why you think "the French" are choosing to make their home in these urban immigrant ghettos. Ordinary French people have been moving primarily to the much wealthier ring of outer suburbs and exurbs further away from the city.

by Bertie on Aug 9, 2011 5:00 pm • linkreport

I like living car free in the city, and I am all for any report that says I'm awesome, but as a statistician, I definitely have questions about this research. I won't get into it here, and I am not suggesting that they didn't check for violations of assumptions, but reporting some standard errors can go a long way in making sure people interpret the results accurately.

by Austin DC on Aug 9, 2011 5:04 pm • linkreport

I did no such thing. I said when calculating the subsidy for comparison purposes you should compare the per trip subsidy, not the per mile subsidy.

Then your position is even more incoherent. What is the point of calculating the per trip subsidy if you don't think the unit of subsidy should be a trip?

But I'm not advocating that. I'm advocating getting rid of all subsidies AND pricing all negative externalities.

Since direct subsidies to transit pay for more than two-thirds of direct costs, getting rid of them means that transit fares would need to triple or more. Adding the cost of negative externalities would increase fares even more. What do you think would happen to the demand for transit if fares were tripled or quadrupled? I thought I was pretty hostile to transit, but it turns out you're even more hostile to it than I am.

by Bertie on Aug 9, 2011 5:10 pm • linkreport

What is the point of calculating the per trip subsidy if you don't think the unit of subsidy should be a trip?

For the fourth time - comparison purposes.

What do you think would happen to the demand for transit if fares were tripled or quadrupled?

as I wrote: "driving would be much more expensive, transit would be somewhat more expensive and walking and biking would remain pretty much the same price as now (free). That would make living in walkable and bikable communities much more competitive. And those communities are the same ones where transit makes sense. So transit would be just fine."

When I got rid of my car, I had to take taxis more often. Taxis are more expensive than driving. So on those trips I lost money, but I made up for it with all the trips where I walked or biked and by not owning a car. Transit would be the same as taxis in my story. An expensive indulgence offset by savings elsewhere.

by David C on Aug 9, 2011 5:19 pm • linkreport

@Bertie,

I'm not sure why you think "the French" are choosing to make their home in these urban immigrant ghettos...

A lot more people have "chosen" to live in these high-rise ghetto towers than in the city proper. Therefore, by your math, they're more desirable. After all, growth maps perfectly to consumer housing preferences.

Let's make an obvious analogy: the cheap seats at a basketball game are better than courtside seats. After all, the number of people who choose to watch from the nosebleed seats dwarfs the number who sit courtside.

Sure, one could argue that the prohibitively high cost of the courtside seats keeps the hoi polloi from purchasing those tickets, but one would be wrong. The high cost of the seats is merely a function of the expense of putting the chairs courtside--I mean, sure the "land" is more expensive, but mostly it's union labor costs involved in unfolding and placing the chairs there. You don't have those costs associated with the more desirable nose-bleed seats, which are self-deploying.

Finally, there's more competition for those courtside seats, unlike in up in the rafters. People don't like endless competition, loud noises, or being sweated on. All of these reasons lead us to the incredible popularity of rafter-seating at professional basketball games.

Did I miss anything?

by oboe on Aug 9, 2011 5:24 pm • linkreport

@Bertie:

The picture you posted appears to be one of the many high-rise public housing projects that have been constructed in the poverty-stricken, crime-ridden inner suburbs of Paris to house the large numbers of poor immigrants that have been arriving from France's former colonies. I'm not sure why you think "the French" are choosing to make their home in these urban immigrant ghettos. Ordinary French people have been moving primarily to the much wealthier ring of outer suburbs and exurbs further away from the city.

It would certainly help your argument if the sprawling poverty in the Parisian suburbs were limited to "inner suburbs", the flashpoint of the 2005 riots in the greater Paris metro area was Clichy-sous-Bois which "is hours away from the city center via public transportation despite only being some fifteen kilometers from the city’s borders."

(http://www.thetransportpolitic.com/2010/01/19/paris-officials-push-huge-suburban-transit-investment-to-increase-metropolitan-mobility/)

For the metrically-challenged, that's about as far out as Tyson's Corner.

Anyway, Paris since the 70s has trended more homogeneously wealthy, and the poor have consolidated in nodes of poverty in suburbia. That's the model that DC's region is trending towards. It's also the reason that transit will likely be a very large part of the transportation mix in the suburbs over the decades to come. The suburbs as a whole will trend more urban. Also more poor. Also more congested.

I suppose one could make the argument that a majority of Americans would prefer the exurban lifestyle all things being equal. But things aren't equal. And as a famous British philosopher once observed, "you can't always get what you want."

by oboe on Aug 9, 2011 5:38 pm • linkreport

I'm done.

Liar.

by oboe on Aug 9, 2011 5:40 pm • linkreport

Liar

Guilty as charged.

by David C on Aug 9, 2011 5:46 pm • linkreport

A lot more people have "chosen" to live in these high-rise ghetto towers than in the city proper.

I have no idea why you think the residents of these high-rise ghetto towers could afford to live in the city proper. Housing in the city of Paris is enormously expensive. That's one reason why almost all of the growth in the Paris metro area over the past 50 years or so has occurred in the suburbs and exurbs. As in America, housing prices are much lower in the suburbs, and it's much easier to get around by car.

by Bertie on Aug 9, 2011 5:51 pm • linkreport

Anyway, Paris since the 70s has trended more homogeneously wealthy, and the poor have consolidated in nodes of poverty in suburbia.

Poor immigrants have consolidated in certain inner suburbs where the French authorities decided to build these hideous high-rise public housing projects. Ordinary, middle-class French people have settled primarily in other suburbs, especially the outer suburbs and exurbs.

That's the model that DC's region is trending towards.

The overwhelming majority of DC's suburban residents are not poor and do not live in high-rise public housing projects. Poor people are not "streaming" into the DC suburbs. Middle-class people are not "streaming" into the District. Yes, the suburbs have become more socioeconomically diverse, and that trend will likely continue. But the vast majority of suburbanites are not poor. In fact, housing values and household incomes in the DC suburbs are substantially higher than the national average. It's a wealthy area with lots of high-paying jobs.

It's also the reason that transit will likely be a very large part of the transportation mix in the suburbs over the decades to come. The suburbs as a whole will trend more urban. Also more poor. Also more congested.

Since most of the growth is occurring in the outer, low-density, car-oriented suburbs and exurbs, it is very unlikely that mass transit will become more important rather than less important to the region's transportation mix, especially given the severe constraints on public spending that are likely to prevail for the forseeable future.

by Bertie on Aug 9, 2011 6:19 pm • linkreport

I have no idea why you think the residents of these high-rise ghetto towers could afford to live in the city proper.Housing in the city of Paris is enormously expensive. That's one reason why almost all of the growth in the Paris metro area over the past 50 years or so has occurred in the suburbs and exurbs.

Sure, but we've already established that the ebb and flow of metropolitan population is a pure expression of consumer preference.

We might just as well say, "I have no idea why you think the residents of [the exurbs] could afford to live in the city proper. Housing in the city of DC is enormously expensive. That's one reason why almost all of the growth in DC metro area over the past 50 years or so has occurred in the suburbs and exurbs."

SYNTHESIS!

by oboe on Aug 9, 2011 6:21 pm • linkreport

For the fourth time - comparison purposes.

But your comparison is meaningless. What is it supposed to mean to compare subsidies per trip when longer trips receive higher subsidies than shorter trips?

as I wrote: "driving would be much more expensive, transit would be somewhat more expensive and walking and biking would remain pretty much the same price as now (free). That would make living in walkable and bikable communities much more competitive. And those communities are the same ones where transit makes sense. So transit would be just fine."

Once again, you are simply ignoring the facts. Direct subsidies account for more than 70 cents of every dollar in direct spending on transit. In 2009, for example, total spending was about $54 billion. Of that total, fares covered covered only about $12 billion. The other $42 billion was subsidies. So if subsidies were eliminated, as you propose, transit fares would have to more than triple to make up for the loss of funding. And on top of that, you want transit users to pay for all the negative externalities of their transit use. The result would be something like a quadrupling of transit fares. And yet you claim that "transit would be just fine." You seriously think that if bus and train fares were increased from $1 to $4 it would have no serious effect on demand, do you?

Transit would be the same as taxis in my story. An expensive indulgence offset by savings elsewhere.

If transit became an "expensive indulgence" there would be much less of it than there is today. There would be massive cuts to transit services.

by Bertie on Aug 9, 2011 6:40 pm • linkreport

We might just as well say, "I have no idea why you think the residents of [the exurbs] could afford to live in the city proper.

They could obviously afford to live in the city proper. As I already reported, the median value of housing units in Loudon County in 2009 was over $500,000. The median household income was $114,000. It is obviously possible to live in DC on an income even much lower than that. But you would get much less for your money. $500,000 will get you a much bigger house in Loudon than in the District. And in Loudon it's much easier to get around by car. The roads in DC are very congested and parking is a nightmare. And Loudon is cleaner and quieter. The public schools are better. There's more room for kids and pets to play. Again, you seem oblivious to the obvious advantages of suburban lifestyles that have been drawing people from cities to suburbs all over the country for 50 years.

by Bertie on Aug 9, 2011 6:51 pm • linkreport

@Bertie,

Surprisingly, finding a house to fit a family with three children in-boundary for a decent school is pretty tough in DC. You may have heard this. Of course, if money were no object, buying a large house near the National Cathedral and sending the kids to private school is quite easy.

Unfortunately, most people in the area can't swing that--certainly not on $115k a year. Living in Loudon is a way to get an approximation of that style of living on the cheap. Just as you can go see the same basketball game Spike Lee is watching--it's just that you'll be in the rafters, and he'll be courtside.

by oboe on Aug 9, 2011 7:20 pm • linkreport

Cleaner environment - Do you really believe that having everyone depend on a car to get everywhere actually is BETTER for the environment? Oy.

Pollution is measured as the concentration of a contaminant in the environment. Higher density tends to mean higher levels of pollution. The more people, homes, motor vehicles, appliances, etc. per unit of area or volume, the higher the level of pollution. This is why Manhattan has some of the worst air pollution in the country. And of course it's not just air pollution. Litter is another form of pollution. And noise. And dirt and grime in general. This is why dense cities tend to be much noisier and dirtier and have much more litter than suburbs.

Less stress - I don't think most people think, "My doctor said I need to reduce my stress - what to do, what to do . . . I know! I'll add a 60 minute commute, and begin and end of my day in virtual gridlock. Ten hours a week in heavy traffic is just the thing to bring my blood pressure down!" To each his own, I guess.

Transit commutes take on average twice as long as commutes by car. And rush hour is when transit vehicles are most likely to be crowded, so you're more likely to have to stand instead of getting a seat. And you get pushed and shoved by other riders as they get on and off the bus or train. I think most people would find long commute in a crowded bus or train pretty stressful.

And it's not just commuting. The noise, the dirt, the crowded sidewalks, the lack of privacy, the congested streets, the puddles or urine and vomit, the homeless people hitting you up for money, and all the other daily unpleasantnesses of big city life also add to the stress.

by Bertie on Aug 9, 2011 7:23 pm • linkreport

You have noted it. You have not linked to this mythical Delucchi study.

There's nothing "mythical" about it. Here's a paper summarizing Delucchi's research. See the table on Page 12. Delucchi estimates total external costs plus subsidies for cars, transit buses, light rail transit and heavy rail transit. His best estimate of the total for gasoline cars is 6.9 cents per vehicle-mile, which is about 4.3 cents per passenger-mile. The best estimate for transit buses is 40 cents per passenger-mile. That's almost ten times the total subsidies and external costs of gasoline autos. For light rail, the estimate is 27-109 cents, and for heavy rail, 17-53 cents. The overwhelming conclusion is that all transit modes are subsidized at a vastly higher rate than automobiles, whether external costs are included or not. As Delucchi puts it:

It really doesn’t matter how one does the accounting. In virtually every case, the total subsidy to transit greatly exceeds the total subsidy to auto use, per passenger mile, in both absolute terms and relative to the prices users currently pay.

by Bertie on Aug 9, 2011 8:51 pm • linkreport

"more desireable" equates to more people wanting it, or, "more people competing for it". Its the same thing.

No it isn't. Higher density means more people per square foot of land, not more people in total. Land prices tend to increase with density because higher density means more people are competing for each square foot, not because higher density is more desirable.

Also -what dcd said about car dependency and the exurbs/suburbs. People are not necessarily choosing car dependency.

Of course they're not choosing "car dependency." If they live in the city, they will most likely have better access to transit, but driving and parking will be more difficult and expensive. If they live in the suburbs, they will most likely have worse access to transit, but driving and parking will be easier and cheaper. Most people prefer to live where driving is easier and cheaper, despite the worse access to transit, because overall a car-oriented transportation mix is just much faster and more convenient than a transit-oriented transportation mix. Some people will choose the transit-based lifestyle instead. People who cannot drive, or don't want to drive, or who have an unusual liking for public transportation or an unusual hostility to cars, or for whom dense urban life has benefits that compensate for the difficulties of car travel. But most people are not like that. The advantages of a suburban, car-based lifestyle are clear to them. Hence the mass exodus from cities to suburbs over the past 50 years.

by Bertie on Aug 9, 2011 9:23 pm • linkreport

Except there isn't an equal amount of transit based neighborhoods to car based neighborhoods out there. So if you live in the car based one (b/c you can't afford the transit one) then you're kind of screwed.

by Canaan on Aug 9, 2011 9:38 pm • linkreport

There are more car-based neighborhoods because most people prefer a car-based lifestyle. Their choices as voters and consumers reflect that preference.

As for not being able to afford to live in a transit-based neighborhood, huge numbers of poor people live in transit-based neighborhoods. You may not be able to afford to live in a nice condo near the Ballston-MU station on the Orange Line in Arlington, but that doesn't mean you can't afford to live in a neighborhood with good transit. You can also use the political process to try and convert your car-oriented neighborhood into a transit-oriented one. You probably won't be successful, though.

by Bertie on Aug 10, 2011 12:03 am • linkreport

Bertie, a few thoughts.

1. Delucchi's analysis seems very suspect to me. He values accidents as practically the same value for heavy rail and driving for instance. Negative. Heavy rail is way safer. This number is one he made up out of whole cloth, based on his 'studied judgement'. He thinks unpriced parking migth be worth 0 cents? Ridiculous. So his numbers do not ring true.

2. He ignores tax subsidization of the oil industry, government program subsidies for cars and driving, safety costs, protection costs for oil production and shipping, subsidies to roads from state, local and federal government, zoning related subsidies etc... So his number is incomplete.

3. The only comparison that matters is the total transporation cost over some set time period based on different living locations. Cost per mile is meaningless unless you account for how your choice of home changes the total miles. If you move 10 miles farther from work, your transportation costs go up, even as your cost per mile remains constant (or even goes down, because parking gets averaged out over the whole trip). The average American drives 15,000 miles per year, and someone who lives in a car-convenient area surely drives more. The IRS puts the cost at driving at 0.51 cents per mile. Delucchi - who's number is deeply flawed and far too low - puts the low end of the subsidy at 5 cents per mile. Note that this doesn't include the cost of parking ones car. This is just the cost of driving it from point A to B. So this number is way, way, way too low. But it still comes out at 0.56 per mile. For an annual cost of $8400 per year.

Last year my wife rode metro to work every day for 250 days a year. That's 500 trips. Metro provides 215 million trips for a total cost (including subsidy) of $1.4B, or $6.51 per trip. That includes the government subsidy. The other externalities come out at less then a penny per passenger mile. My wife only goes 5 miles to work, but I'll throw a whole dollar onto the price per trip to cover the negative externalities. It's now $7.51. All her other trips are on foot. Her total transportation costs are thus $1877.50 per year. Even if she takes a $20 taxi four times a week, her cost only goes up to $4037. That includes all subsidies and externalities. And how is that possible? The convenience of living in a dense urban neighborhood.

Now, I realize my wife may not be average among urban dwellers. A lot ot people walk to work, so there costs are even lower. I bike to work, so my cost is probably half hers. So she's not representative.

So yeah, if you get rid of subsidies and making walking and biking far cheaper by comparison, I think transit and urban living is going to be just fine.

by David C on Aug 10, 2011 12:08 am • linkreport

...so their costs...

by David C on Aug 10, 2011 12:08 am • linkreport

So, in other words, a two-unit condo is more expensive to build than a couple of single-family homes, irregardless of land costs.

Suburban housing tends to be one-story or two-story. Dense urban housing tends to be midrise or higher. It's generally more expensive (higher construction costs) to build vertically than horizontally. Suburban housing can use inexpensive materials. Wood or block frames, concrete stucco or siding, tile or shingle roofs, etc. It doesn't need much skilled labor or special equipment. It doesn't need expensive one-off architectural designs. It doesn't need to comply with strict fire codes or health and safety laws. And in single family homes, the entire structure is private living space.

Dense urban housing, especially high-rise housing, generally requires more expensive materials. Lots of steel and reinforced concrete. More skilled labor and equipment. More expensive designs. It has to comply with stricter fire codes to facilitate evacuation and fire suppression. More complex plumbing to move water and waste between the floors. And much of the space in multi-unit buildings must be given over to common areas to allow access to the private residencies -- lobbies, hallways, stairwells, elevator shafts. All these requirements add to the costs of construction.

by Bertie on Aug 10, 2011 12:52 am • linkreport

He values accidents as practically the same value for heavy rail and driving for instance. Negative. Heavy rail is way safer.

The numbers do not reflect the "safety" of each mode but the EXTERNAL costs of accidents. Most of the costs of automobile accidents are internalized through insurance.

He thinks unpriced parking migth be worth 0 cents? Ridiculous.

"Ridiculous" is not an argument. On-road parking is covered by the cost of building the road. In fact, drivers pay for metered on-road parking twice -- once through the taxes and fees they pay to construct the road, and again through the parking meter fees. Most off-road parking is private, and most unpriced off-road parking is bundled. Businesses include the cost of the parking they provide to their customers in the prices of their goods and services. It's not an external cost. Public off-road parking is rarely unpriced. In any case, parking costs are a quibble. They don't alter the fact that transit is subsidized at a vastly higher rate than autos.

He ignores tax subsidization of the oil industry, government program subsidies for cars and driving, safety costs, protection costs for oil production and shipping, subsidies to roads from state, local and federal government, zoning related subsidies etc...

More quibbles. Either the subsidies you mention are trivial, or they apply to both transit and automobiles, meaning that they don't offset the enormously higher direct subsidies provided to transit, or both. Or they are incomprehensible (I have no idea what "safety costs" is supposed to mean).

The only comparison that matters is the total transporation cost over some set time period based on different living locations. Cost per mile is meaningless unless you account for how your choice of home changes the total miles. If you move 10 miles farther from work, your transportation costs go up, even as your cost per mile remains constant (or even goes down, because parking gets averaged out over the whole trip). The average American drives 15,000 miles per year, and someone who lives in a car-convenient area surely drives more. The IRS puts the cost at driving at 0.51 cents per mile. Delucchi - who's number is deeply flawed and far too low - puts the low end of the subsidy at 5 cents per mile. Note that this doesn't include the cost of parking ones car. This is just the cost of driving it from point A to B. So this number is way, way, way too low. But it still comes out at 0.56 per mile. For an annual cost of $8400 per year.

Another paragraph that is utterly confused. Yes, the longer the commute, the higher the costs, both internal and external. So what? What does that have to do with Delucchi's estimates? Where does your "0.56 per mile" come from? And what's it supposed to be? The external cost of driving? The total cost? Or what? And how does the fact that the IRS estimate for driving costs excludes parking mean that it is "way, way, way too low?" And what does THAT have to do with Delucchi's estimates of external costs? You seem incaplable of making any kind of coherent argument. Just a series of disconnected assertions that don't seem to have any relevance to the issue under discussion. Many of your comments are like this.

So yeah, if you get rid of subsidies and making walking and biking far cheaper by comparison, I think transit and urban living is going to be just fine.

You are living in a fantasy world. Getting rid of transit subsidies would require a quadrupling of fares to cover the lost funding and maintain the same level of service, assuming no loss of ridership. But of course if fares quadrupled there would be a huge loss of ridership. Which means the remaining riders would have to pay even higher fares. Which would mean an even greater loss of riders. And so on. It's a vicious circle. The only plausible outcome of eliminating transit subsidies is a massive reduction in the size of the nation's transit system. The only transit services that would survive, if any, are the ones that provided a benefit worth the market price of the ride.

by Bertie on Aug 10, 2011 1:46 am • linkreport

There are more car-based neighborhoods because most people prefer a car-based lifestyle. Their choices as voters and consumers reflect that preference.

I have always found it difficult to reconcile the phenomenon of "slugging" as a reflection of consumer preferences. Seems like a real pain in the ass to me, but there's no accounting for taste.

Vive le difference!

by oboe on Aug 10, 2011 9:34 am • linkreport

Most of the costs of automobile accidents are internalized through insurance.

Not even close. Many insurance policies are capped. And crash damage often exceed caps. Especially if there is a loss of life. The FRA values human life at $10,000,000 for example. So additional costs are absorbed elsewhere. And something like 40,000 people die every year in road fatalities caused by cars. So, not even close.

"Ridiculous" is not an argument.

It is when everyone knows that 0 is a completely insane number.

Businesses include the cost of the parking they provide to their customers in the prices of their goods and services.

YES!!!! OMG Yes!!!!! And that is an external cost charged to everyone whether they drive or not. You can't really be this unaware, right? This is an act. How is adding cost to everything everyone buys to serve only those people who drive not an external cost?

Either the subsidies you mention are trivial

See my link above. They are not trivial

or they apply to both transit and automobiles

Not to the same degree. Automobiles are almost 100% reliant on oil. But transit is nowhere near that. It is probably majority electric, and to a lesser degree CNG. And many of these costs are related to oil. Others are significantly lower per mile for electricity.

Regardless, we agree that he left these out and that therefore that (plus that he left out the government subsidies to driving - which you didn't address) makes his "analysis" incomplete and thus irrelevant.

I have no idea what "safety costs" is supposed to mean

There are a lot of things you have no idea about. It's in the study I linked to way up in the comments.

Yes, the longer the commute, the higher the costs, both internal and external. So what?

Living in the suburbs creates a longer commute. Which is why per mile isn't the right measure.

Where does your "0.56 per mile" come from?

As I wrote "The IRS puts the cost at driving at 0.51 cents per mile. Delucchi - who's number is deeply flawed and far too low - puts the low end of the subsidy at 5 cents per mile." Add those two together. You can do that right?

And what's it supposed to be? The external cost of driving? The total cost? Or what?

Obviously it is the total cost.

And how does the fact that the IRS estimate for driving costs excludes parking mean that it is "way, way, way too low?"

Because it isn't including all the costs. I'm calling it total while acknowledging that it isn't.

And what does THAT have to do with Delucchi's estimates of external costs?

Why do you think point three is about Delucchi's esitmates of external costs. It is a separate point, which is why I separated it.

You seem incaplable of making any kind of coherent argument.

And you seem incapable of understanding anything beyong an over-simplified mantra. This is a nuanced, possibly even counter-intuitive argument. Try to keep up.

You are living in a fantasy world.

My fantasy world has almost no transit because that would take space away from the bikini models.

Getting rid of transit subsidies would require a quadrupling of fares to cover the lost funding and maintain the same level of service, assuming no loss of ridership.

Absolutely 100% false. Wmata's subsidy (I linked to it) is about 30% of cost. That's how much transit goes up.

The only plausible outcome of eliminating transit subsidies is a massive reduction in the size of the nation's transit system.

Not if coupled with an elimination of both direct and indirect subsidies from driving.

Let me try to simplify it for you. Look, don't try to attack the numbers first. Just try to understand the concept. Then, when you understand the concept, you can debate the numbers.

1. You seem to think that it is a choice between driving 40 miles a day OR taking transit 40 miles a day. It isn't. It's a choice between driving 40 miles a day OR taking transit 6 miles a day and walking 3 miles a day.

2. Making driving and transit more expensive, makes biking and walking cheaper in comparison.

3. Biking and walking are really ony an option in the city.

3. Living in the city means relying on transit.

4. While the cost of transit would go up - perhaps even more per trip than driving - it is only one part of your transportation diet.

5. The daily cost of transporation in the city would increase by the increased per mileage costof transit(X)*6 miles

6. The transportation cost associated with living in the suburbs, meanwhile, goes up for every mile of transportation.

7. The daily cost of transporation in the suburbs would increase by the increased per mileage cost of driving(Y)*40 miles

8. The ratio of average miles driven to average miles of transit is 40/6, which is equal to 6.6

Do you understand all of those ideas?

So if X/Y is less than 6.666 then removing subsidies makes urban transportation cheaper relative to suburban transportation.

Even using Delucchi's estimates (which I think are skewed toward driving) puts the ratio for transit bus at 33/5 which is 6.6. Light rail is 27/5 which is 5.4. And Heavy rail is even better at 17/5 which is 3.4. In every case, daily suburban transportation costs goes up more than daily urban transportation costs. Same is true if you use the high end of the range.

The only transit services that would survive, if any, are the ones that provided a benefit worth the market price of the ride.

That's not quite right. The transit services would survive are the ones that make a cheaper lifestyle possible. And that would be almost all of them, as well as ones that would be built later.

Also, I'm unsubscribing from comments, so I won't see your reply.

by David C on Aug 10, 2011 9:58 am • linkreport

More about the explosion of poverty in the suburbs--or as Bertie calls it, "everyone choosing the suburbs over the city":

http://www.tnr.com/blog/the-avenue/93338/desperately-seeking-bloomberg-the-%E2%80%98burbs

There are two main reasons the suburbs have been successful over the last 40-50 years: first, there's the Ponzi scheme effect, where we push further and further outward, bringing in new revenues to pay for old infrastructure. As the burbs age, the sheer costs of maintaining the sprawling fabric become untenable. The only way to avoid that is to keep pushing outward: Chevy Chase, then Aspen Hill, then Gaithersburg, then Burke, then Fredrick, etc... As the new areas come "online", the old-growth suburbs decay. Growth moves outward.

Secondly, the suburbs have benefited from a comparative lack of poverty, which carries a heavy financial cost in addition to social costs. Little poverty means less policing resources. Less social services. Heck, less sanitation services--as poor people tend to litter more than middle-class people. It goes from top to bottom.

The explosion of suburban poverty (compared to the urban core, which has trended in the opposite direction) means the costs of suburbia will increase as well.

Even if we concede that some folks like living in brand spanking new suburban cul-de-sacs with excellent services, and convenient access to job centers, those places are becoming in shorter and shorter supply as the "new" developments move ever outward.

Moving corporate campuses further outward doesn't help, because while you're moving the corporate headquarters closer to some subset of suburbanites, you're simultaneously moving it further away from everyone else. So, no, Oracle and the tech corridor are not moving to Leesburg anytime soon. As population increases ("more people are choosing the suburbs!"), and congestion gets worse, it's much more likely that they'll move further inward--either into the city proper, or to one of the coalescing urban-like centers like the proposed Tyson's Corner.

And the value-add of such centers will be density and transit.

by oboe on Aug 10, 2011 11:30 am • linkreport

A whole other factor, that makes this more complex, is that transit systems (because of their federal subsidy) must meet a host of government requirements that include things like "Buy America" mandates and an onerous environmental review process. So while getting rid of the subsidies will raise the cost to passengers it won't be by the whole current cost of this system, because the cost of the system will go down.

If drivers had to "Buy America" their costs would go up too.

by David C on Aug 10, 2011 12:33 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.

or