In the last post I pointed to a set of data out of the City of Portland showing that in a 12-year period when the city added four new rail transit lines, including the globally marketed Portland Streetcar, the percentage of Portland residents who take transit to work (called transit’s “journey-to-work mode share”) seems not to have changed at all.
Commenter Pantheon dug into Canadian statistics and found that in the period 1996-2006, when Portland’s transit journey-to-work mode share was idling in the 12-15% range, Canadian cities posted these gains:
Ottawa up 2.2% to 19.4%
Montreal up 1.2% to 21.4%
Calgary up 3% to 15.6%
Vancouver up 2.2% to 16.5%
To be fair, these aren’t that much higher than Portland’s 12-15%, but it would be nice to think that Portland would show better outcomes than Calgary.
I suspect that parking costs are a big issue, especially since Portland is still a relatively easy city to drive in. My colleague Stuart Donovan offered this comment:
I suspect Portland’s lack of headway on mode shift reflects subsidies for parking, rather than any failing of transit investment or urban land use policies.
Colliers study of international parking rates found Portland ranked near cheapest in the U.S. – at just $9/day. This is comparable with Louisville, KY – a city which is half as dense.
Investment in transit is likely to be futile in an environment of subsidised parking. Removing minimum parking requirements and applying a parking levy would be a good start.
Parking is one of the major costs associated with a private vehicle based transport system, yet many cities around the world persist with ill-considered regulations which force developers to provide excessive amounts of parking.
Instead of futile debates on the merits of BRT versus LRT, we would be better off advocating for parking reforms, especially abandoning minimum parking requirements, as a way of achieving better transport outcomes.
So I went and looked at parking costs in Portland’s downtown. Stuart’s right. At the 10th & Yamhill garage, which is right on two rail lines and a streetcar line, and just a few blocks from the very center of downtown, you can still park all day for $10. The highest SmartPark rate anywhere in Portland is $15/day at 1st & Jefferson. Meanwhile, in Calgary, a day’s parking can cost up to C$22 (US$21.28) in the City Centre and Convention Centre parking structures.
Stuart goes on:
Given that the average car-park takes up about 380 sq ft a “market” price for parking is closer to $20/day, rather than $10/day. I smell parking subsidies.
Not to say that they should immediately double the cost of parking, but there’s obviously room to drive considerable mode shift from parking policies alone.
In other recent work we’ve been doing, we’ve repeatedly seen that parking price is the most powerful locally-controlled lever for shifting people out of single-occupant cars, in the absence of more direct congestion charges. Increases in parking costs drive big shifts to transit or other options.
During the heady early years of Portland’s urban revolution, in the 1970s-80s, Portland had a cap on the supply of parking downtown. Short-term parking was always available, but all-day parking was supposed to get expensive, so that commuters would have a motive to use the transit system or other sustainable modes. The cap was lifted in the 1990s, and if 10th & Yamhill is any indication, it looks like there has been no price pressure since then; the price today isn’t much higher than it was when my office was near there 15 years ago.
Since 10th & Yamhill is a city-operated garage, the city may be subsidizing parking (by charging a below-market price) in conflict with its own sustainable transport goals. If that’s true, it wouldn’t be unusual. In fact, most cities subsidize parking more than they would like to, but politically it’s very hard for a city to let parking costs rise to a true market rate.
Cities that have no control over their parking supply, ironically, may have an advantage on this crucial sustainability-driver, because their private providers can let all-day rates rise to the true market level (also causing transit ridership to rise) and the city government can rightly claim that there’s nothing they can do about it. It’s just that beautiful free market, doing it’s thing.
The parking issue is real: The cost to drive should never be perceived as just the cost of gas. Drivers should not only understand their costs of depreciation, maintenance, and repair…but should also be paying their fair share of rent for the valuable downtown space that their cars take up. I think cities should be raising parking prices, but they should also be doing marketing efforts to get drivers to understand the true costs of driving. Seriously, most drivers don’t understand that driving in downtown traffic can raise your cost per mile of driving by 3x or more.
Land use in Portland is also a huge issue. Planners don’t typically tell us directly what they are thinking when they plan sections of street, but we can tell what they are thinking by how they zone that street. In portland, the planners have spoken loud and clear: They think people only work and sleep…and maybe a few of them go to school.
In an area where you have the amazing utility of a frequent service streetcar or light rail, you need so much more to be able to persuade people from their cars. You need entertainment: Movies, live theaters, clubs, bars, and concert halls. You need different types of shopping: grocery, drugstore, clothing, electronics, housewares. You need restaurants. You need public functions like libraries, police stations, utilities offices.
I’m to the point where I think planning and zoning has served its purpose and now needs to go away. Planners can’t ever possibly understand everything that people need along a rail line in order to make that line infinitely useful. Take a look at every single high ridership line in the US, and you should be able to notice one thing: Every single building was built before zoning and planning came into effect. The low ridership of newer lines is a reflection of the rapid-transit-utility-killing effect that imperfect zoners and planners have.
In your criticism of “planners”, what part of town are you referring to? Much new development around transit lines is mixed-use, not exclusive-residential. Of course, there is lots of exclusive-residential stuff still going up–there is a market for it, after all, and many lenders are leery of financing construction of anything else–but it would be incorrect to suggest that mixed-use development is being ignored or discouraged along transit.
Portland, though, has a more fundamental problem–we’ve got plenty of retail and entertainment options around town; what the city and metro need more of is design, manufacturing, etc–the stuff you can build an economy around.
There is a role for zoning and planning, obviously–it is very useful to buy a home and know that a hog farm won’t (legally) move in next door; certain things like heavy industry and commercial agriculture are segregated from living space for good reasons. OTOH, I’ll agree that the separation of residential from retail/office has been largely disasterous–but that’s not a reason to abandon zoning altogether–if that’s what you are suggesting.
Has someone looked into train frequencies? I have never lived in or visited Portland myself, but from what I’ve read, I’ve always been astounded for all their invested energy in rails, the MAX and streetcar doesn’t run more than once every 12-15 minutes for most of the day (a quick look at the Blue Line schedule shows a 15-minute headway at 6pm!)
By contrast, the Vancouver Skytrain seems to run at least every 10 minutes all day long (obviously with even shorter headways during busier times).
If I were to make choices about where to live and what mode of travel to use, I just don’t think a 15-minute headway train would have nearly as much lure as systems that can be relied on more as a walk-up service.
Be careful what you wish for. In the short-term, a sharp increase in central city parking prices might induce a significant shift from autos to transit. But over the long-term, the primary effect may be to drive businesses and people out of the city and into the suburbs where land is cheaper and parking can be provided at lower cost.
One difference between SkyTrain and MAX is that the former is a driverless metro–adding additional trains is cheap; whereas MAX needs a driver. Operatinal funding is a challenge for TriMet (a problem which is in many ways the fault of the agency, but still).
Man oh man, if the cost of parking is the biggest driver of transit use, then I hope Chicago’s recent privatization of the parking meters and the corresponding quadrupling of rates will have a positive impact on transit use (and hopefully transit expansion). Do you know how expensive does parking have to get before this happens? Is there a tipping point at a certain price level?
@Watson, Bingo. That’s the essential tension at play and the prime reason why downtown retail and real-estate interests continue to politically (and in some cases financially) support subsidized parking garages.
My sense is the city parking garages exists to attract shoppers as part of Portland’s successful revitalization of downtown.
@Adam. I hope Stuart chimes in here with some references, but my view is that when we’re just focused on the impact of parking prices, there’s no particular tipping point; every increment of price makes a difference to transit mode share, at least up to the level of the market price for parking, which most US cities are well below.
@Thom. Yes, frequency is always a big deal. To be fair, most of Portland’s segment of the main LRT line has Blue and Red lines combined, yielding frequencies no worse than 7.5 min all day.
But Calgary’s rail system is similar to Portland’s, though it’s a little more frequent. Calgary LRT is every 10 minutes all day on each line, for a net 5 min through the urban core. (A larger share of Portland is at the better two-line frequency than is the case in Calgary.)
@Christopher. We’re talking here about all-day parking rates, which don’t affect shoppers. It’s normal for rates to be much lower for 4 hours or less. Parking for shopping may be one of the public anxieties that helps justify subsidies for parking in general, but the subsidies we’re discussing here affect commuters, not shoppers.
One fantastic tool to encourage transit use that is used in Sydney is to levy a tax on commercial parking spaces in Central Business Districts, both in the centre of the city and at suburban centres well served by public transport. The annual fee is currently AU$2000 in central Sydney, which is about AU$8 per weekday (US$7). The funds are used to build and maintain public transport and park-n-ride facilities throughout the city.
Thanks for this post, Jarrett.
As a newcomer to Portland, I’ve been surprised that I haven’t seen a single parking space in Southeast that one can actually pay for. It’s all free (although I am guessing there may be some exceptions close in and close to Burnside). I was unaware that parking prices in the downtown were also lower than one might expect in a city of this size and density.
Given the impact on the quality of infill development and incentives for using different modes, parking policy should be a critical topic for the Portland Plan being updated right now.
For “Vancouver up 2.2% to 16.5%”, this is actually for the entire region, the transit mode share for the City of Vancouver is 25%.
SkyTrain peak headways are only 108 seconds on the busiest sections of the line. For non-peak periods, the headways are 5-8 minutes. It is really nice not to have to wait long for the trains. You never have to look at a schedule.
Daily rates for some private lots in Calgary are now pushing $30/day. Also, Calgary’s LRT lines run on a 5-min frequency on AM/PM peaks and 10-min freq. on off-peak periods during the day and in the evenings until 10pm. Combined frequency in the City Centre at peaks is about every 2-3 minutes.
You’ll find parking meters and such on the campus of PCC Southeast Center, at 82nd and Division–I suspect that other colleges in town charge for parking as well. (Portland Community College, which has three main campuses in the Portland area as well as a couple more smaller campuses, has free–for students and staff–hourly shuttle connecting the campuses; an element of the Portland system sometimes overlooked).
Donald Shoup has written a very detailed book on the costs of free parking provision, which also looks at this issue: http://www.amazon.com/gp/search?index=books&linkCode=qs&keywords=1884829988
Stuart’s colleague in New Zealand, Julie Anne-Genter, has also done a study into parking subsidies and how they impact transportation: http://www.nzsses.auckland.ac.nz/Conference/2008/papers/Genter.pdf
Julie and Stuart also star in this really interesting video on parking subsidies: http://www.youtube.com/watch?v=l_O6dR7YfvM
Watson: on the contrary, the combination of transit and high parking fees tends to reinforce the strength of downtown. Even the most comprehensive transit networks have strong hub-and-spoke elements, which means they both help and are helped by monocentrism.
In practice, you can see this in action in New York and Chicago, where the comprehensive transit systems have kept those cities downtown-dominated while other downtowns lost ground to edge cities. Calgary has seen the same: Calgary already had a strong concentration of work downtown before light rail opened, but in the almost 30 years it’s had light rail and parking restrictions downtown has if anything grown stronger.
Almost all old transit-oriented central cities have either lost population or grown much more slowly than their surrounding suburbs over the past few decades. To what extent this is due to the cost and difficulty of driving in central cities, and specifically the cost and difficulty of parking, isn’t clear. But there’s no basis for the claim that the long-term effect of more costly parking is to attract businesses and people to central cities rather than drive them away.
A lot, Alon, depends on the prestige and quality of the downtown. Were some Manhattan firm, in a bizarro universe, to threaten to relocate its head office to New Jersey unless more parking for private autos were provided by the city, they’d be laughed out of the room–it wouldn’t be a credible threat, due to the tremendous amount of prestige Manhattan enjoys. (And given that the NY/NJ/CT metro area enjoys excellent transit links, being a large, dense, and yes, rich city; such a threat wouldn’t be necessary in the first place).
Portland’s downtown is highly desirable, but doesn’t have the same cachet–many business here have made similar threats here WRT land use policies, and some have been carried out. Portland’s lot is better than some industrial cities, where downtown consists of little more than high-rise office buildings, only occupied during working hours, surrounded by slums–but even Portland is still an auto-dominated city.
We’re just less auto-dominated than other US cities of comparable size. 🙂
The real “solution” would be to bulldoze all the freeways–which obviously is not going to happen anytime soon.
The City of Portland made a decision to susidize short-term parking in downtown to foster business and especially support downtown retail, that might otherwise have moved to suburban malls with lots of free parking, in the 1980s and 1990s.
The Brookings Institute had an interesting study about job sprawl.
One of the points the study made was that old transit orientated cities like Chicago had some of the worst job sprawl in the country. 17.9% of all jobs are within 3 miles of the CBD in Chicago. 15.5% of all jobs are within 3 miles of the CBD in Philly.
Now compare that with places known for sprawl like Phx or Vegas. In Phx, 25.8% of all jobs are with in 3 miles of the CBD. In Vegas 29.9% of all jobs are with in 3 miles of the CBD.
(see table 2)
If you look at the areas that had the bigget problems with job growth decentralization between 1998 and 2006, it has included a lot of the cities most publicly committed to adopting smart growth, with Portland, Sacramento, San Jose and Salt Lake Cities on the Brookings list of cities under going most rapid decentralization. (see table 5)
Smart growth has been effective at encouraging urban gentrification of city centers by wealthier and better educated residents to move back downtown.
The employment growth that is occuring in downtown Portland is largely a non-market activity relying on universities and government.
But Watsons point about higher parking rates driving out employers is spot on.
Portland also has a low $1.25/hr subsidized rate for parking under 4 hours in the Downtown area to stimulate local business. With the new light rail and streetcar lines, their transit system should be good/appealing enough that they could charge closer to a market rate and be okay. It is weird though–Portland does has a surprising amount of surface parking lots and big, cheap parking garages.
Actually, you don’t want all the jobs in the “city center”, unless that’s where everybody lives as well. One particularly bad model is where everyone travels downtown to work, then back home to the ‘burbs to live.
And having a healthy reverse commute (as Portland does–a major employment center is in the “Silicon Forest” of Washington County, which draws many workers from the city) helps transit be more efficient. Westside MAX runs full in both directions during the morning and evening rush hours.
Many jobs are best suited (for mobility purposes) close to where people live. Those jobs which must be centralized benefit from clustering, as transit can be optimized to reach them–but not too much clustering, as then capacity issues crop up and travel times can increase.
It seems to me that in cities where it would be desirable to locate a business, the parking is always expensive. And in hollowed-out cities parking is always cheap. But it’s hard to sort out cause and effect.
In any case, it would be incorrect to assume that greater population growth in suburbs is due to consumer preference regarding lifestyle. Consumers react to market prices, and market prices for real estate are often determined by government regulations.
In Canada the reverse of what you describe has happened. Vancouver now has a higher residential density than Manhattan. But why? Let me tell you a story.
For a long time Toronto was following the model of sprawl, but in the 1990’s the city council made a decision that it wanted to encourage density. So what did they do? They changed the height regulations on condominium buildings in the city, which had always been absurdly low. This caused controversy because 50-story towers started going up in older neighbourhoods and they would cast shadows over houses. Irate neighbours complained that these behemoths were blocking out the sun. But the ultimate effect is that
a)builders can now make a profit out of land that was previously not profitable to buid on, which means
b) they build, and build big, and finally
c) greater supply of housing in the city pushes prices down and makes city living affordable.
The result is that there is now more population growth in Toronto than the surrounding suburbs, a reversal of prior trends.
If regulations prevent huge condo towers in cities (i.e. density) then of course there will be more growth in the suburbs, because houses in the city will be expensive due to the artificially low supply that results from overbearing government regulations. So don’t assume that greater growth in the suburbs is due to real free market decisions.
It’s funny how conservatives are all for allowing the free market to do its thing, except when it comes to parking or real estate or anything else that affects them personally.
Government regulations are created through the democratic process. I am skeptical of the claim that regulation has been a significant cause of suburban growth and central city decline. But whatever its role, regulation is an expression of political preferences. Democracy isn’t perfect, but suburbanization has been going on all over the country for half a century. And in Europe too. If people didn’t want pro-suburbanization regulations, they would have changed them. You mentioned one reason why people support regulations that limit density: aesthetics. People generally don’t want their communities to become concrete jungles where high-rise buildings block out the sun. Other negative effects of density include congestion, crowding, noise, pollution, crime and loss of privacy.
The overwhelmingly dominant trend in the United States and Europe for many decades has been the growth of suburbs and the relative decline (in many cases, an absolute decline) of central cities. I haven’t looked into Canada specifically, but I would be very surprised if it has not also followed this pattern.
” we would be better off advocating for parking reforms, especially abandoning minimum parking requirements,”
Portland eliminated downtown parking minimums in 1995, and replaced them with maximum entitlements. Banks and lenders require parking, not the city.
The result is that there is now more population growth in Toronto than the surrounding suburbs, a reversal of prior trends.
According to Wikipedia’s entry on Toronto, citing data from Statistics Canada, there has been no reversal, at least through 2006. Between 2001 and 2006, for example, the population of the city of Toronto grew by a little more than 20,000. The population of the Toronto Census Metropolitan Area grew by over 400,000. And the population of the Greater Toronto Area grew by almost 500,000. The overwhelming majority of recent population growth in Toronto has been in the suburbs.
The most livable and desirable density to me comes in the form of cities like Amsterdam where buildings are 4-5 stories high and you have plenty of transportation options: tram, subway, walking, it’s safe to bicycle, and lots of parks & non-motorized streets & squares. And the canals…it’s just my ideal setting. You still have larger apartment buildings, but not so many towers over 10 or 15 stories. In minutes you can cycle to the outer edge of the city and still get anywhere you want safely by bike and neighborhoods feel connected, yet open at the same time. You can have a house and a yard if you like and still have plenty of transport options – even by car if you desire. That’s the kind of city I want to live in.
Democracy doesn’t work so well on issues where a small group has a strong interest, and the majority mostly doesn’t care one way or the other, or only has a fairly weak interest in the issue. This is exactly what goes on with government-subsidized suburban sprawl: the population at large doesn’t care much, while land speculators stand to profit handsomely, and thus have an incentive to get politicians to pass regulations favorable to them. Which of course they did, and boy are there a lot of those, starting from subsidies to mortgages and the FHA, which promoted suburban homeownership over urban renting, and continuing on to Urban Renewal, wherein the government bulldozed urban neighborhoods to turn the land over to developers.
And as for “regulations are created through the democratic process”, well, the current financial crisis is to a significant extent a result of changes in regulations (repealing various restrictions passed after the Depression), and those changes were also a result of “the democratic process”, as were the much-hated bailouts that followed.
Democracy doesn’t work so well on issues where a small group has a strong interest, and the majority mostly doesn’t care one way or the other, or only has a fairly weak interest in the issue.
I don’t see how this follows. It seems to me that public policy should reflect not merely how many people feel one way or another about an issue, but also how strongly they feel it. Democracy is more than just counting heads. And I would dispute the idea that our land-use and transportation policies reflect the will only of a passionate minority rather than how most people prefer to live and get around.
In any case, what’s the alternative? Democracy isn’t perfect, just better than all the alternatives. And as Arthur Schlesinger likes to say, one of the greatest strengths of democracy is its capacity for self-correction.
I don’t see how this follows. It seems to me that public policy should reflect not merely how many people feel one way or another about an issue, but also how strongly they feel it.
To the extent that those who don’t care can abstain, yes. But you make the mistake of assuming that “strong feelings”, or more accurately put, commitment, can be measured in dollars or political power. The Tea Party republicans and the netroots progressives probably have equally strong “feelings” on a good many issues; however, the former seems to enjoy significant financial backing and other forms of institutional support that the latter does not. How much money (instead of votes) should be permitted to talk in politics is an interesting debate.
Democracy is more than just counting heads. And I would dispute the idea that our land-use and transportation policies reflect the will only of a passionate minority rather than how most people prefer to live and get around.
Democracy in the US is frequently about counting dollars, unfortunately. And the petrochemical and auto industries enjoy much more political clout than do the railroads and train manufacturers. Those who have lots of dollars seem keen on keeping it that way.
At any rate, you seem to be committing the Fallacy of Democratic Inertia (don’t google, I just made it up)–asserting that the status quo in a democracy represents the Will of the People, thus it should not be challenged. As you note, a great strength of democracy is self-correction. I suspect the “will of most people”–those who aren’t really passionate about transit issues and just want to get where they are going–depends highly on the price of gas.
I didn’t say you shouldn’t challenge the status quo if you think it’s wrong. Of course you should challenge it if you think that. Proponents of higher density and more transit are challenging it. The point is that suburbanization and car-oriented development is not some recent or isolated phenomenon. It’s been happening for at least half a century throughout the United States and Europe. Yes, democracy is imperfect. Markets are imperfect. But it is unlikely, to say the least, that this trend could have persisted for so long and in so many places if it were fundamentally inconsistent with how most people prefer to live and travel.
Perhaps preferences have recently changed, or are in the process of changing, to favor a large-scale shift to higher density and more transit. But I see no serious evidence of any such shift. Just a lot of wishful thinking supported by anecdote and poor reasoning.
There are many, many, many different reasons why people choose to live where they live:
* Amenities they prefer; some prefer the cosmopolitan scene, some prefer huntin’ and fishin’; some just want cheap rent and a good fast Internet connection.
* Quality and cost of (non-mobility) public services such as schools
* Proximity to employment–both in terms of living in a city where jobs in their profession are plentiful, and in terms of having a short commute.
* Desire (or lack thereof) for communal/non-communal living, large living spaces, backyards.
* Proximity to other amentities and services
* Explicit mobility desires–a desire to live where cars, or transit, or walking, or bicycling, is convenient
* Proximity to (or distance from!) family and friends
* Proximity to persons of similar demographic characteristics and/or desire to avoid those seen as undesirable (there’s a reason the term “white flight” was coined).
* Cost of housing and of other goods and services
* Availability of housing.
As to how much these factors drive an individual (or a “typical individual” I’ve no idea). But any assertion that people live in the ‘burbs because they want to maintain an auto-dependent lifestyle is dubious without some evidence–it may well be the reverse is true; someone wants to live there because it’s cheaper than a downtown condo, or because they work at an office park nearby, or because there’s fewer brown people, or because they want separate bedrooms for the kids–and guess what, you need a car to make it work.
The evidence of a general preference for suburbs and cars over density and transit is half a century of consumer and political choices concerning where to live and how to get around by hundreds of millions of people throughout the United States and Europe. If that is not evidence of people’s preferences on these matters, what would be?
In fact, sprawl is not merely an American and European phenomenon, but a global one. It seems to be the inevitable result when a democratic society becomes wealthy enough for mass ownership of automobiles, except where it is precluded by geography, as in Japan and Singapore, for example. (And Singapore has a rather autocratic form of democracy).
Thanks to Watson for so politely pressing his case. My view on this is that suburban car-oriented life is so heavily subsidized that nobody knows how many people would freely choose it if it weren’t. There’s no question, though, that older, less car-oriented inner cities are repopulating over much of North America and Australasia.
I think different people like different things. If everyone paid the true costs of their choices, there would be no objection to anyone living in suburbia who wants to. Even without that levelling of the playing field, cities with well-redeveloped cores, like Portland, are at least presenting a clear alternative to suburbia, as opposed to places like Las Vegas where the core is so small and unloved that suburbia is basically the only option.
We also know that urban density and its amenities (including transit) drive land values very high, raising concerns about affordability. From the free-market perspective, high prices mean that something is very desirable, and that we should be making more of it.
I’d like to see your evidence that suburban car-oriented life is as heavily subsidized as you suggest. How did these subsidies arise, and how have they managed to endure for so long and in so many places, if they do not reflect what people really want? Public policies in democratic societies are not imposed on an unwilling population by unaccountable overlords. They are created and sustained by the collective political choices of the people.
I’d also like to see your evidence that “older, less car-oriented inner cities are repopulating over much of North America and Australasia.” The data I have seen on domestic migration shows an unambiguous long-term trend of outmigration from central cities into suburbs, and from old, dense transit-oriented cities to new, sprawling, car-oriented ones. The rate of loss has slowed in the United States in recent years because of the recession and housing bust, but there’s no sign of a reversal. Some old, dense cities, such as New York, have gained population even though they’ve had a large net loss of domestic migrants, but that’s almost entirely because of immigration. Many cities are attractive destinations for new immigrants because they have large existing immigrant populations, relatively generous public services (including transit), and lots of low-level service jobs. That doesn’t make them very attractive destinations for most people already living in the U.S. though.
And high prices do not mean that something is very desirable. Housing costs tend to rise with density not because denser housing is more desired but because it costs more to supply. Land costs are higher and construction costs are higher, so the market clearing price is higher. If there were more demand for denser housing, builders would be happy to supply it.
Watson, to take your last point, when you say “land costs are higher” for high density, you’re making my point. Zoning for high density raises land value, because land zoned for high density and located around public transport is scarce. If there were more rapid transit, there would be more developable land around transit, and you would see that this price softening to the point that more affordable high-density housing would get built.
I will let readers form their own judgments about whether dense inner cities are repopulating. I can only cite the development energy that everyone can see in Vancouver, Portland, Seattle, San Francisco, Oakland, Los Angeles, Minneapolis, Sydney, Melbourne, and Brisbane, to cite only cities with which I’m pretty intimately familiar. All of these cities are seeing pressure to convert underutilized core industrial areas to residential, and are filling in other available spaces with housing ranging from townhouses to highrises.
Re subsidies to car travel, and thus to suburban development forms that depend on cars, see this post:
$10 all day? We pay more than that in *ITHACA*, which is a fraction the size!
“The point is that suburbanization and car-oriented development is not some recent or isolated phenomenon. It’s been happening for at least half a century throughout the United States and Europe. ”
In the world of building construction, where buildings built 500 years ago are still in use, that *is* a recent and isolated phenomenon.
The longer-term trend is a shift out of rural areas and into cities, which has been happening for, oh, at least 200 years. People who like suburbs are people who wanted to move to the cities but still get some of the benefits of rural living.
The only serious study I’ve ever seen estimated that nowadays 40% of the population wanted to live in suburban development, 40% wanted to live in true urban areas, and 20% wanted to live in genuinely rural areas. And this was a very rough estimate. The “brownstone” or “row house” style of living — where you *do* get your own yard, but not a huge lot — is not really accounted for, and its popularity far outstrips availability right now.
Current laws have prevented supply of housing from matching demand in the US — suburbs are oversupplied, superdense urbs are undersupplied, “medium density” is badly undersupplied, and rural is either getting eaten by suburbs, or is impossible to make a living in. :-/
Watson is assuming that things are the way they are because people wanted and decided for them to be that way. This is a false premise. Many things are the way they are because of historical accident or artifact. New Orleans can’t elect not to have hurricanes. How many people REALLY want MS Windows?
Many aspects of suburbia were sheer circumstance. Minimum parking requirements with prescribed levels for each enterprise was rather arbitrary. Other cities adapted monkey see-monkey do. The result was huge, sprawling land use. Many people prefer suburbia b/c it’s the only thing they know, like my Windows analogy. As mentioned above, it’s inertia to a larger degree than preference for the modern suburban aesthetic.
Land costs are higher for high-density housing because of the nature of supply and demand. Higher population density by definition means that more people are competing for each square foot of land. Higher demand per unit of supply means a higher unit price. This effect has nothing to do with zoning. It’s simply a consequence of market pricing.
You’re right that zoning laws may further increase the price of land in dense areas, but that’s a different mechanism. Land zoned for high density is rare because high density has high negative externalities – congestion, crowding, noise, pollution, and loss of privacy, among others. Each additional person in an area of land increases these costs for everyone else.
Another negative externality of higher density is loss of aesthetic value. New York could increase its housing density if it allowed developers to build on the land occupied by Central Park or Washington Square. San Francisco could increase its density if it allowed developers to tear down all its beautiful low-rise historical buildings and replace them with high-rise condominium towers. The residents of those cities understandably oppose such measures. People value beauty, historical character, public spaces. They don’t want to live in a concrete jungle. Even New Yorkers.
So people have perfectly rational reasons for supporting zoning laws that limit density and restrict development in other ways. Different people have different ideas about the appropriate kinds and degrees of restriction, and zoning laws emerge from the collective expression of these preferences through the political process.
This doesn’t make sense:
You’re right that zoning laws may further increase the price of land in dense areas, but that’s a different mechanism. Land zoned for high density is rare because high density has high negative externalities – congestion, crowding, noise, pollution, and loss of privacy, among others. Each additional person in an area of land increases these costs for everyone else.
In other words, high-density costs more because its unpleasant to live in? How does THAT work?
It is probably worthwhile to distinguish between upscale and downscale high-density. Many slums are high-density, and are both cheap to live in and highly undesirable–having the unpleasantness of crowds, an abundance of crime and other pathologies often associated with poverty–and few of the amenities one would desire. (Transit is often provided, but other attractive enterprises, particularly private ones, prefer not to locate there).
Upscale density–the Pearl District in Portland, Central in Hong Kong, most of Manhattan–has crowds, but crime is generally lower and there’s a host of amenities to choose from. Such places are often the most expensive to live in BECAUSE they are desired; there’s a reason it costs more to live in Manhattan than New Jersey.
Certainly, simply increasing residential density without providing other amenities is unwise–a community of residential high-rises devoid of shopping or unemployment would be a transit nightmare. It’s the mixture of things that is important, and it’s the lack of mixture which makes many suburbs problematic, where you have to get in the car to buy a loaf of bread.
And yes, among the amenities it is useful to include, even in high-density areas, are things like parks and such.
Here is a very recent analysis of domestic migration at the state level over the period 2000-2009, which cites the Census Bureau as its data source. As you can see, out-migration is dominated by northeastern states with old, dense transit-oriented cities, plus California. In-migration is dominated by southern and western states with new, sprawly, car-oriented cities.
And here is an analysis of domestic migration at the county level since 2001. Again, it shows an unambiguous trend of migration out of central (“core”) counties and into suburban counties. Yes, some core counties have posted a net gain in domestic migrants. But most of them have suffered a net loss. Where total population has grown in core counties, it has come almost entirely from immigration of people from other countries.
I don’t see how this data can be reconciled with the idea that there is any meaningful “back to the city” trend in American demographics. Some people are moving back to the city, and many cities are gentrifying parts of their dense historical cores, but there is no general trend towards high-density, transit-oriented lifestyles. The general trend is the reverse, as it has been for decades.
In other words, high-density costs more because its unpleasant to live in? How does THAT work?
No, the supply of land for high-density development is limited by zoning laws that restrict density. That limitation on supply increases prices. The reason people support laws that restrict density is that density has the negative externalities I listed. I don’t think this is very controversial. If you ask people why they want to limit development in their community they will usually express concerns about congestion, crowding, noise, privacy, aesthetics, etc.
Now that makes more sense. Zoning laws in many places RESTRICT density, making (nice) high-density regions scarce.
In some locales, there is a scarcity (i.e. none) of nice high-density development, the only sort of high-density to be found are slums–a state of affairs which frequently causes suburbanites to a) think lesser of high density, and b) restrain such with zoning laws.
Regarding migration patterns; much of this has to do with the deterioriation of the US industrial base, I suspect, as opposed to people leaving Pittsburgh for Phoenix because they want larger lot sizes…
My stand corrected. Toronto’s suburbs continue to outpace the city in the growth numbers. My statement to the contrary was based on my memory of reading newspaper articles a few years ago that said the trends were reversing. Still, the condo boom in Toronto is real and I imagine the numbers would be worse without it. Also, note that some places counted as outside the Toronto CMA are in fact beginning to resemble urban areas in their own right. I am thinking of Richmond Hill, a densifying city of high-rise condos in York region, which is the fastest growing region in Toronto (23% over 5 years). They plan to extend the subway out there in the near future, and much of the building boom there is no doubt predicated on that assumption. So just remember that in a city like Toronto, areas classified as suburban may not always be what you are picturing. I suppose suburbia ain’t what it used to be.
Jarrett, one more comment of yours I want to address:
Re subsidies to car travel, and thus to suburban development forms that depend on cars, see this post: …
Considering the small value of the benefit (a tiny fraction of the total cost of driving), it is hard to see how parking subsidies could have the huge effect on lifestyles you seem to be attributing to them. In Europe, driving is far more expensive than in the U.S. Gasoline prices are two to four times the price in the U.S. Taxes and fees are also higher. Road congestion is worse. And yet cars overwhelmingly dominate the transportation system in Europe too, just not quite as much as they do in the U.S. Europe has been suburbanizing like crazy over the past 50 years, just like the U.S. For economic, historical and geographical reasons, Europe will probably never become as sprawly as the U.S., but it has clearly been following the same basic trajectory.
And again, if these subsidies are fundamentally at odds with how people want to live and get around, why haven’t we eliminated them? Comments by others suggest that parking subsidies, far from being unpopular, may serve purposes that people value highly and that would not be served through pure market pricing, such as the protection of downtown business neighborhoods.
And as I’ve mentioned elsewhere, transit in the United States is subsidized over 70 cents on the dollar in direct subsidies alone. If transit users had to pay the full costs of their rides, fares would almost quadruple. That subsidy seems to me a more likely source of distortion in consumer choices than the small parking subsidies for drivers.
If I couldn’t find occasional downtown parking for less than $11 or $12, I would exclusively take the Max. This is not so much a testimony for public transit as a declaration that I could not personally justify more than about $7 (for surface parking) to $11-12 for a full day in a garage. As a young creative professional, I will sacrifice walking a few blocks to save a few bucks on parking.
As it stands, I take the Max as often as possible. One problem, however, is the west side park and rides fill up EARLY. I have had less than a 50% success rate finding a spot at Sunset, which forces me to park downtown, or plan ahead and leave earlier or further west on the line.
Do you live in Aloha? 🙂 If not, where on the westside do you live? Unfortunately, the bus service connecting TO Max on the westside is spotty (only one frequent service bus line, the 57/TV Highway, in Washington County–it feeds into MAX and needs a transfer at Beaverton TC). If taking the bus to MAX isn’t an option, the Millikan Way station doesn’t fill up quite as fast as Sunset TC–and if you get desperate, there are several office parks within walking distance that have ample parking and frankly don’t care if you use their lots as park-and-rides. (Some of them will tow you though).
Beaverton is an interesting case–its average density is actually higher than Portland, although it’s mostly denser-than-average suburbia with numerous apartment buildings (and trailer parks) sprinkled about, plus a few smatterings of TOD along the MAX line. There are a few corridors which probably could justify decent bus service based on density and proximity to employment–but other than the aforementioned TV Highway line, there isn’t any frequent service. Given that much of the population does own autos, non-frequent service isn’t likely to attract any non-transit-dependent riders; but a frequent line or two might…
@Aloha. Your problem with being unable to find parking at MAX light rail stations in the west is, of course, evidence that these spaces, too, are underpriced. The same principle applies to Park-and-Ride: You're using 12-15 square meters of real estate for an entire day, and in a city such real estate is never truly free. You should expect that as downtown parking prices push beyond $15, you'll need to start paying $1-2 to park at the suburban stations.
Watson, the suburbs have been outgrowing the central areas for 200 years in fast-growing areas (slow-growing or depopulated ones sometimes have infill – see downtown areas in e.g. New York, London, Chicago, Philadelphia). Even then, when there was no room in the central city, the city expanded out. Before the 1920s those suburbs were exclusively streetcar-oriented, but the principle was the same. It wasn’t parking that did anything in.
It’s not that “the effect of parking isn’t clear” – that phrase of yours is pure FUD. It’s that the effect is clear, and zero. Parking problems are not officially used as reasons for relocation, unlike climate or taxes or regulations, and they’re not correlated with decline.
And if anything, postwar US cities that bent over backward to accommodate drivers crashed harder. On average, each freeway through an urban core contributed to a population decline of 16%. Not surprisingly, the only two old cities to have surpassed their 1950s population peaks, New York and San Francisco, were early adopters of freeway revolts.
Employment is fleeing the urban core. Chevron left SF for San Ramon. Bank of America left SF for North Carolina. The venture capital firms left SF for Sand Hill Road in the very suburban Palo Alto.
In the bay area, the Fortune 500 companies are now mostly found in the Silicon Valley, not in SF.
In the bay area the largest billing office most large accounting and law firms is in Santa Clara, not SF.
Economically, SF is now a bedroom community for the pennisula. Fortune 500 companies want to be near the airport but are quite indifferent to actually being inside the City of SF.
In the 1800’s travel on land was expensive and travel by boat was comparatively cheap, so employers and people concentrated in dense cities. When innovations like elevators and steel construction allowed increasing densities, Manhattan and the loop in Chicago got even denser.
But during the last century the cost of transportation on land dropped dramatically. Employers no longer need to be near ports. Until the 1960’s Ghiradelli Chocolate was made in SF. Today its made in San Leandro. Its old factory in SF is a shopping center catering to tourists.
High land transportation costs no longer drive employers to central business districts.
Employers want to be near high quality labor forces, but they don’t have to be in old city business districts. If you look at the high wage employment in the bay area, the Googles of the world aren’t in dense cities but went for spread out suburban office campuses instead.
The most recent revival of urban cores isn’t employer driven. There is no wage premium paid by employers in central business districts. Its being driven by employees.
Chevron had no problem leaving SF. But some employees want the amenities of urban life. They want the cafes and better restaurants found in cities. Incomes are lower in SF than Santa Clara.
But housing prices are similiar or higher in SF vs Santa Clara. In short people are paying a greater share of there income for the privledge of living in a city.
The point that Watson made is spot on. If you keep driving up the cost for employers in being central business districts, they will flee. They are already fleeing.
Mike, your comment consists of anecdotes, all in support of one general factoid:
Unfortunately, this factoid is incorrect.
SF is a net employment gainer. The Census Bureau’s county to county commute statistics readily show that SF gains 77,000 commuters a day from San Mateo and Santa Clara Counties and sends 59,000. SF’s balance with the East Bay is 121,500 to 25,500.
Housing prices are similar in SF vs. Santa Clara, but office space rents are not. Parking is a rounding error by the standards of what it costs to rent an office in a high-rise on Market Street. To say nothing of the fact that the majority of SF’s CBD’s employees take transit anyway…
We are talking about the change in marginal employment growth. Employment growth in the bay area is in the suburban areas. The agglomerative economies that formerly supported the creation and expansion of the central business district in SF have completely broken down. Fortune 500 companies no longer feel they need to be located inside the City of SF. The legacy fortune 500 companies in SF are leaving, new ones aren’t relocating to the central business district in SF. Even the existing fortune 500 companies that are staying in SF, aren’t staying in central business district (Levi’s left the CBD for an office campus in SF, Industrial Light and Magic got space in the Presido). The financial, insurance, real estate industries that formerly used to cluster in dense central business districts near corporate headquarters also no longer see the need to be downtown.
When you have lost the investment bankers and venture capital firms, you have lost the last most conservative firms that previously felt they needed a downtown address. In the SF Bay Area, you have already lost them (they went to Sand Hill Road).
Buildings are a long duration asset. Because they formerly served an economic purpose, in the past there was plenty of high rise office construction in SF and that created an abundant supply of office space in the SF central business district. But SF really isn’t building much new office space anymore.
The tall buildings that it is building are residential (not office) projects, like Rincon Towers. At the same time some of the existing older office space is being converted into residential uses.
Employment growth has been fastest in Silicon Valley, but it’s also been fast in the central city. The growing bedroom communities in the Bay Area are the East Bay and the counties further out in the metro area, such as Marin and Sonoma.
The only data I have here that’s more recent than 2000 is the BEA’s numbers on inflows and outflows of commuter earnings, and the balance. Inflow measures the total earnings of people living in the given county and working elsewhere, outflow measures the opposite, and balance equals inflow minus outflow.
According to this measure, San Francisco and Santa Clara are both job centers: their balances are strongly negative. San Francisco’s balance used to be the more negative of the two, but Santa Clara’s balance became more negative at the end the 1990s’ tech boom; following the tech bust, SF and Santa Clara’s balances have been about equal, and grown at the same rate.
The BEA’s job numbers only go up to 2000, but even then, SF is growing as a job center. It has fewer commuters from the Peninsula than it used to, but it’s more than making up the difference with people who both live and work in the city and with commuters from the East Bay. In general, even in the 1970s, population decline in the central cities happened at the same time as job growth.
Despite the buzz about edge cities and reverse commuting, the suburban areas that have a negative balance are those that developed as independent metro areas. Silicon Valley developed in the early 20th century, as Stanford encouraged professors to patent inventions and start their own companies. It only joined with San Francisco to form one metro area later. The other suburbs, to the north and east of San Francisco, have large net positive balances; even Alameda County has only a neutral balance.
“Employment growth has been fastest in Silicon Valley”
Its not just that employment growth has been growing faster in the Silicon Valley. It has also been growing on the Pennisula and on the inner Bay Area communities like Oakland, Berkeley, Emeryville etc as well as along the 580 corridor, Concord, Walnut Creek, and Pleasanton Its just not showing up in the data because because your data sources aren’t fined grained enough to break down the data, by political subdivision inside these counties. The bedroom communities would be Tracy, Dublin, Pittsburgh, Antioch, Vacaville, Fairfield as well as Marin, and Napa. Because counties like Contra Costa and Alameda have some cities that function mostly as bedroom communities and others that are employment centers, the aggregation makes them seem like bedroom communities.
The Silicon Valley, and inner East Bay looks a lot more like Los Angeles than Manhattan. All of it is car oriented and highly suburban in nature.
Part of the reason the high value knowledge work is bleeding out the CBD is for reasons that Gladwell wrote about here. Employers today are turning away from office spaces in CBD’s. They just aren’t effective places to work. They have much better luck with the more open corporate campus.
But the other reason is the agglomerative economy in the bay area is the region, not as much the city of SF.
If you want to to biotech, you are going to do that in the East Bay in Berkeley, Emeryville or in the areas of Oakland closest to Berkeley or Emeryville. This is where the researchers from Berkeley set up there companies like Chiron.
The Silicon Valley functions as a one stop place for setting up a high tech firm. You have the accountants, the lawyers and vc’s who know about the requirements to do an IPO. You also have plenty of engineering talent willing to switch firms to support a good idea.
If you look at the new companies that are being started its been a long time since an idea got formed in the CBD of SF. Pixar was developed in Emeryville. Industrial Light and Magic got set up in a ranch in Marin County. When it moved to SF it went to the Presido, where it got its own office campus there. But what company got its start in the CBD of SF?
Companies want access to the labor pool of the bay area, but they have no specific need to be in SF, especially not in the CBD of SF. The employment growth that SF is having is in the former warehouse district near the new UCSF campus. Again this growth is more corporate campus in nature, its not in the financial district and its similiar to the growth in Emeryville, Berkeley or the Pennisula.
The fortune 500 companies/major employers are located through out the region. Safeway is headquarted in Pleasanton. Kaiser and Oracle have big operations in Pleasanton. The department of Energy’s Joint Genome Institute is headquartered in Walnut Creek. Chevron is now headquartered in San Ramon.
There was a time when all of that work probably would have been done in CBD of SF. Today none of it is done in the CBD of SF. Employment in the bay area is all over the place.
High housing prices force people to move further from the core to buy a house. The commuters improve the qualify of local labor force in the area, the employers noticing the stronger labor force and cheap rent follow and the core of the bay area thus expands.
The problem for the CBD of San Franisco is that employers are just or more happy in San Ramon, Emeryville or San Mateo as financial district in SF. This is why the office towers in the financial district are going residential and the new towers that are being built like the Rincon Tower are residential.
I don’t disagree with most of what you’re saying. I’m arguing two points: a) San Francisco employment may be in decline relative to various edge cities, but is in absolute growth, and b) parking rates are such a tiny percentage of the cost of doing business that they’re unlikely to trigger employer flight.
A bit of data from the trenches: a colleague and I conducted an informal survey of public transport riders in April of 2009, on the 3 month old Phoeniz, Arizona, USA light rail, asking them “What’s the main reason you are taking the train?”. To make a long story short, the first two people (rather ordinary types at that) we asked told us the main reason they were taking the rather successful new Phoenix light rail to downtown was because of the cost of parking. It wasn’t to save energy, or improve the environment, or save time, or safety, etc etc. They took the train to avoid the cost of downtown parking.
We were quite surprised, in fact I’ve got a copy of Donald Shoup’s book about parking and I was still surprised at how zoning regulations requiring parking in new construction, and subsidized parking in America in general are in some ways wrecking public transport.
If raising parking rates would have only a tiny effect on the cost of doing business for downtown employers, then it is likely to have only a tiny effect on driving/transit commute shares. Most employers or employees would likely just absorb the higher cost.
Regarding the issue of employment decentralization in the Bay Area, Brookings published a study on job sprawl last year. The authors placed the San Francisco metro area in the “Most Decentralized” category for large employment centers, with over 57% of jobs more than 10 miles from the CBD. They also found that San Francisco is experiencing “moderate decentralization.” Between 1998 and 2006, the share of metro area jobs within 3 miles of the CBD declined by 2.6%. The share of jobs more than 10 miles from the CBD rose by 2.0%.
For the nation as a whole, the authors also found a clear trend of job decentralization. Between 1998 and 2006, 95 out of the 98 metro areas studied were found to have a decrease in the share of jobs located within 3 miles of downtown. The absolute number of jobs increased, but job growth in the outer-most parts of the metro areas was much higher than job growth in downtowns (17% growth in outer areas compared to less than 1% growth in downtowns).
Watson. A statistic of the form "% of jobs over 10 miles from the CBD" says nothing about the viability of transit, which is this blog's ultimate concern. What matters to me is how many jobs are near good transit without much parking, as opposed to in business parks by the freeway. The Bay Area contains many nodes of dense employment around transit — downtown Oakland and Berkeley, for example — and while these would be counted as decentralization by the metric you mention, they are actually evidence of a desire for urban lifestyle and amenity built around a high quality transit service. A single downtown concentration of employment is not actually the ideal development pattern for transit, for reasons I explored here:
The best market for transit is a constellation of dense centres with two way demand flowing between them. If you wanted to build the perfect transit metropolis in the Bay Area, you'd put more job growth in towers around stations (with minimal parking) in Oakland and Berkeley and other East Bay nodes, even though that would count as "decentralization" by the metric you suggest.
Yes, an outward flow of jobs from a single dense center to smaller surrounding dense centers would be measured as an increase in decentralization by Brookings’ metric. But I don’t think that’s what’s been happening. The jobs aren’t moving from a big central CBD to smaller satellite CBDs. They’re moving to suburban office parks, industrial parks, shopping malls, schools, hospitals, etc.
Do you have a blog of your own? I’d love to follow it.
First, parking can be a large percent of commuting cost, which can influence a commuter’s choice of mode of transportation. However, it’s not a large share of total cost of living, so it doesn’t do much to increase the cost of employment. Again, look at Calgary, whose downtown has only grown in importance in the 30 years since the light rail regime started.
Second, big downtowns are likely to remain big. In fact, Robert Lang and Jennifer LeFurgy have found that after edge city formation in the 1970s and 80s, in the 90s most major US downtowns actually increased their share of the total metro area office space, including San Francisco (though I believe the study looks at MSAs, which means Silicon Valley is excluded). The downtown share increased in 7 out of 13 metro areas, and decreased in only 3.
The Brookings trend for more dispersed employment is partly a trend for larger downtowns, spread over more urban neighborhoods. For example, the Lang/DeFurgy study points to Jersey City’s renaissance in the 1990s: it’s basically an extension of Lower Manhattan. It’s more than 3 miles from Midtown, but it’s not job sprawl. (In general, New York is and remains the most CBD-dominated metro area in the US, and had a growth in CBD office space share in the 1990s. Apparently, Manhattan parking doesn’t make people flee to suburbia.)
Amenities downtown matter.
For some businesses, a “prestigious” address, usually meaning a downtown address in a first-tier city, is something they will pay extra for. For others, not so much.
But the quality of a downtown, and its ability to draw, is based more on how many people want to LIVE there, not merely work there or shop there. Plenty of cities have downtowns full of commuters in the day, and then nobody but the homeless at night. That’s not the model you want to emulate.
Pdx has a “mature” mass transit system; it’s been around a long time. in that same period you mention, bicycle commuting has grown greatly. our population has grown. so between losses to bicycling (significant) and simply adding new bodies to the region, Pdx holding steady indicates strong use of the system.
granted, we need to make parking/driving less easy & more expensive, by TriMet is still doing very well — for an agency run by people with zero business sense.
Informative blog and thoughtful discussion. Is this really the internet?
I live in Birmingham, United Kingdom, and the cost of parking (which in my case would’ve been deducted from my salary) was equivalent to the cost of public transport, which I elected to use as it has a travel-time advantage. However, this is largely due to malice-aforethought on my part when choosing where to live, offsetting the more expensive property close to the train station against the purchase and running costs of a (second) car in the first place.
My wife’s work is mutually exclusive with mine from the point of view of usefully fast and reliable public transport, so as a couple we retain the advantage of a car for long suburban-to-suburban journeys (often with luggage) and shopping. Noticeably, parking charges do not prevent us driving into the city for some few hours for entertainment purposes. Drink-driving responsibly and congestion (especially at Christmastime) are much more significant barriers. This last implies my wife and I would benefit from Birmingham introducing congestion charging (which seems to fill the political space of the parking-charging debate here). We’d either value the premium service of congestion-free motoring or the improved public transport service the charge would subsidise.
Yes, and for that same reason, it can influence a worker’s choice of where to get a job. High downtown parking prices are an incentive to work in the suburbs instead, where parking is cheaper. And that in turn can influence employers’ choices of where to locate their workplaces. Transit proponents hope that increasing the price of downtown parking will induce people who would otherwise drive downtown to use transit instead. But over time it might also, or instead, induce people to work or shop in the suburbs instead of downtown. That’s why, as others have said, subsidized parking remains politically popular among downtown businesses.
But most of the job sprawl Brookings found is more than 10 miles from the CBD. Jersey City is much less than 10 miles from midtown and lower Manhattan. For the New York metro area, Brookings found a trend of “moderate decentralization” between 1998 and 2006.
The numbers here don’t match up. Within the choice of how to commute, parking costs can be significant. Within the choice of where to work, there are more decisions about cost of living than just transportation, which is on average 12% of the cost of living in the US.
It’s this factor of 8 difference that makes one a significant issue and the other a triviality.
It’s well-documented in New York that businesses overestimate how many people drive to them. They scream at various pedestrian/bike improvements that, it turns out, have no adverse effect on business life.
The authoritative reference here, I believe, is Fischel’s paper explaining that NIMBYism is social insurance against change that possibly makes neighborhoods less livable. Fischel does not talk about business NIMBYism, but the principle is the same – businesses are not compensated if a public project backfires and reduces profits. So they stick with what they know, regardless of whether it works or not.
This doesn’t happen so much lately – Manhattan’s outgrowing Westchester, Fairfield, and Middlesex. The trend that people in the region do talk about is reverse commuting – people who live in New York City and commute to the edge cities, often on transit.
For what it’s worth, transit mode share in the region is going up, not down. The same is true in Calgary, which has North America’s second highest parking rates. The fact that both the metro areas with the highest downtown parking rates are seeing an increase in transit share should dispel the idea that free parking is a requisite of strong transit-oriented downtowns.
I just thought about it more… is there any evidence that, to the extent decentralization is actually occurring, it has anything to do with parking?
For example, have businesses cited it as a reason for leaving? (No – they usually cite land costs and taxes.) Have growing edge cities cited free parking as an advantage? (No – they again cite land costs and taxes, and sometimes talk about not having urban problems as a euphemism for “We’re in the favored quarter and have few poor people and minorities.”)
If you look at the CBD of SF the two tallest buildings were office building built during the late 60’s, early 70’s. The Transamerica Building was built in 1972, the Bank of America building was built in 1969. But in the remaining almost 40 years, it hasn’t managed to build any office buildings taller than that.
The reason this is significant is that Market Street has the best transit access in SF. It has Bart, Muni and most bus routes in the city end up eventually going through Market Street. In short if there is an area where transit should be driving infill employment growth it should be the SF financial district.
If you look at the tall buildings built during this decade in SF, they were residential towers, Rincon Tower and the Millennium Tower. The last tall office towers built in the CBD were done in the mid 80’s (almost 25 years ago).
For employers, the SF financial district is losing its Mojo. The financial district is reinventing itself as a residential community. Its converting some of the older office space into residential units and its been building new residential towers. But its appeal for employers is falling. I think a big part of that is the lack of parking.
If you look at where the employment growth in SF has been occuring its has been near the UCSF Mission Bay Campus and China basin. If you google those areas, you will notice these areas have some of the largest concentrations of off street parking in the entire City of San Francisco.
As the economies of agglomeration for employers in the financial district have fallen apart, employers have moved to places with more parking. (Think Chevron to San Ramon). Even with transit connections as strong as strong financial district of SF, employment growth in SF is still following the parking lots, out to China Basin and Mission Bay.
If you look at the broader bay area, you also see that pattern. While some neighborhoods in Oakland and Berkeley have excellent transit connections. The employment growth has been in the areas nearest Emeryville.
Bart doesn’t go through Emeryville. Amtrak does have a station in Emeryville, but that station serves tourists from Sacramento catching a bus to and from San Francisco. Amtrak doesn’t sell tickets on its buses from SF to just Emeryville, nor does it offer a monthly pass from SF to Emeryville. In short for local commuters, Amtrak’s Emeryville station really isn’t an option and its not pricing its tickets to be a viable option for commuters. If you want to argue that Amtrak should reverse its policy, I agree, but I also want to point out, that employment growth is again being directed by proximity to the freeway and the availability of off street parking in the staging areas of the former warehouses, not by access to quality transit.
What is noticeable is that we aren’t seeing new office towers/employment centers built in the areas with excellent transit access in the East Bay. We aren’t seeing new big employers relocating to Shattuck Ave in downtown Berkeley, adjacent to the Bart Station. We aren’t seeing office projects near the West Oakland, or Ashby Bart Stations.
Instead employment growth seems to be following the freeways and the parking lots.
The successful transit oriented developments in the bay area have all been housing driven, not employment driven. At Fruitvale station, they added more housing. At West Oakland Station, they added more housing. In downtown Berkeley, both the university and non university actors built more housing.
Lastly if you look at transits share of the commute in SF, it is still down since at least 1990.
Lower land costs allow you to build parking lots. So when they cite lower land costs, that is another way of saying that they are leaving for areas with better parking.
How relevant are land costs when zoning codes permit construction of skyscrapers? Total cost of construction is probably more relevant for a business which owns its office space.
Many businesses don’t own, however; they rent (whether downtown or in an office park). How do office rents in SF compare with comparable rents in the Valley or elsewhere in the Bay Area?
Mike, the employers could say “parking,” but they don’t. They do say “freeway access” sometimes, so it’s not some conspiracy not to mention roads. Lower land costs’ primary benefit isn’t parking, but space for server farms and large office complexes.
As for San Francisco’s tallest buildings: Manhattan’s tallest building opened in 1931. New York is still monocentric around Manhattan, and this if anything has increased in the last 20 years, once people stopped being afraid of city streets.
The residential aspect of TOD isn’t really accurate. What’s more accurate is that TOD office construction tends to be more mixed-use – see e.g. recent construction in Lower Manhattan. This looks residential by the standards of the single-use office complexes that have arisen in the edge cities, but it’s not; those downtowns are still dominated by commercial uses, just less than they used to be in the urban renewal era.
Parking is definitely a reason why some businesses leave downtown in my home city, but it is not the reason used most of the time.
You know why many employers like to leave? Keeping up with the Jones’ They like the nice, new shiny office space, with all the new amenities, whatever the latest definition of “Class A” office space might be. They also do this if they’re expanding, which can be easier in a new office building than existing ones where the vacant space is fragmented. Odds are the new buildings will have at least some parking. Old CBD’s undoubtedly have old buildings from before the auto era. The new office space, all things being equal in our perpetual growth economy will not be downtown because the downtown is likely to be heavily built out, or at least limit in new construction sites. The outer areas are more likely to be subject to modern suburban zoning that requires obtuse amounts of parking whether tenants or developers want that much parking or not.
Some of this argument above is too confounded to conclude anything meaningful from. Obviously, once a CBD develops to a certain level, it becomes difficult to develop much more. (Largely due to assembling contiguous properties.) As a metropolis grows in population and land consumption (sprawling or not), the amount of places one could develop other than the CBD increases.
Political issues of condemnation are a driver of this. Once said CBD becomes dense, the surrounding areas are likely to have developed into residential or otherwise cherished property. Property owners may not be interested in selling out their neighborhood to see the CBD expand. Therefore, a new BD will emerge detached and away from the CBD.
As one can see, arguing for a mankind preference for decentralization cannot be argued using growth of satellite business districts versus the orignal business district in a metropolis.
CroMagnon: what you say about opposition to CBD expansion is equally true for edge cities – hence, the formation of edgeless cities. In fact, it’s easier to do it with edge cities, which typically form in politically connected favored quarters. CBDs were surrounded by poor and working-class neighborhoods until 20 years ago; urban renewal was basically a way of extending a single-use CBD further out into the rest of the city.
There are some possible dilemma Portland is facing if parking fees are to be raised:
1) Businesses can easily move to nearby business community in or around the city for cheaper or free parking. Worse they would move out of Portland/state. Vancouver WA USA is located across the Columbia River.
2)parking spaces glut? Force parking operators keep fees low to attract customers.
3)Some businesses have their own attached parking garage or own parking lot next door. May not affect employees at all if city parking fees raised.
4)more people would use transit, less parking revenue the city will get. Peeps are not that stupid.
5)lack of major tourist attractions in downtown.
6)not enough city-operated garages?
7)city economy is not that strong.
When Chevron left SF for San Ramon, the explicitly sited parking as one the reasons they left. If you look at their new headquarters on Bollinger Canyon Road, you will notice its surrounded by acres of free parking. More than half the land at there new headquarters is parking lots. This argument that they are moving for more server space just doesn’t square with the empirical data. Look at what they are doing with the new land. Its parking lots.
Part of the appeal of the former warehouses is the abundant parking in those locations.
Remember even in the City of SF the share of people commuting by transit is only 31.9%. In the overall bay area region, the share of commuting by transit is 9.8%.
The reason employers are relocating out of the financial district to areas with parking is that parking lots give you access to the 90% of the employment pool in the bay area who has poor access to transit near their homes along potential commute routes. When an employer has parking near the place of employment it expands the pool of potential applicants and expands it labor pool.
The residential preponderance of transit oriented developments may not be accurate in New York or Manhattan, I don’t live in Manhattan, but it is true in the bay area. To argue otherwise just reveals your ignorance of the local facts on the ground in the bay area.
In the bay area the agglomerative economics of clustering are breaking down for employers, but they are strengthening for residences. If you live in a high density neighborhood, it can support more restaurants and cafes open later. As crime rates have fallen in inner cities like SF, Berkeley and Oakland, living in inner cities is less dangerous, this has made urban housing more appealing in these locations. This is what is driving the success of the housing component of transit oriented developments along the bart stations.
The continued presence of many more fortune 500 companies in Manhattan may be sustaining the agglomerative employment economies in Manhattan in a way that is no longer happening in SF. My understanding is that the investment banks haven’t departed Manhattan for Greenwich Connecticut quite to the extent that the investment bankers and venture capitalists departed for Sand Hill Road in Palo Alto. My understanding is that in terms of the major financial insurance and real estate firms, Manhattan is still overwhelmingly the dominant office in the New York region. So in that sense, Manhattan may be a data outlier here.
If local employers in Portland are citing parking as one of the reasons they are relocating, I believe them. There was a time when SF functioned closer to the Manhattan model than it does today. But the forces that pulled apart the appeal of the SF financial district for employers in the bay area probably are working even stronger in Portland. Compared to SF, there are even fewer firms headquartered in Portland so the CBD in Portland is probably much weaker.
Speaking of parking “subsidies”…
In Portland, Westside Light Rail (the Blue Line to Beaverton and Hillsboro) is disproportionately used by park-and-riders; those who drive to a TriMet owned and operated parking lot (where parking is free for up to 48 hours) and ride the obiviously subsidized transit system.
Likewise, there are major parking garages at the Gresham, Clackamas Town Center and Gateway Transit Centers.
As a bus rider who lives in a suburb but walks to an unimproved bus stop to reach downtown, I’m being forced to pay for this luxury (and there are two major park-and-ride lots on my bus route) through increased fares and decreasing service, while other folks get subsidized parking paid for through transit fares. Fair? Hardly.
And yes, Portland is unlike most cities in that Portland has a significant public parking agency which owns most of the downtown parking garages, while in most cities parking is near, or entirely privatized.
Can you substantiate this claim?
Here is the latest ridership data from APTA, showing the change in unlinked trips for New York-area transit systems from January to September 2009:
MTA New York City Transit: -3.78%
MTA Staten Island Railway: -6.87%
MTA Long Island Rail Road: -6.52%
MTA Metro-North Railroad: -4.35%
MTA New York City Transit: -2.88%
New Jersey Transit Corp: -4.69%
Keep in mind that these ridership declines occurred before the recent “doomsday” cuts to New York transit, which seem likely to reduce ridership even more.
While I won’t argue that park-and-rides are underpriced (a $1 or $2 surcharge might help TriMet with budget issues, and still is cheaper than parking downtown even if one were to pay full-fare rather than use a pass)–the notion that THEY’RE (Max riders) being subsidized but YOU are not, is simply wrong.
No transit mode operated by TriMet has a farebox recovery ratio > 100%. MAX is closest, busses are generally far lower. A bus to the suburbs, depending on where, is most likely a low ridership route (I don’t know which line you use and where you live), and as such is probably heavily subsidized.
While I do share concerns about TriMet expanding its network without having the operating funds to do so; I disagree with the notion that certain modes (i.e. local bus service) or routes (especially older ones) are somehow entitled to first cut at operational dollars. Jarrett did a post which discussed the distinction between “coverage” and “ridership” services (MAX is definitely the latter, a suburban bus line is likely the former), which you should read.
BTW–haven’t seen you on portlandtransport in quite a while… miss sparring with you over there.
Yes, but the longer-term trend reported in that study is a substantial decline in downtown’s share of new office space and a substantial rise in “Edge Cities”‘ and “Edgeless Cities”‘ shares. The authors state that their data is more consistent with the “decentrist view” that Edgeless Cities will come to dominate. Edgeless Cities are the lowest-density of the four urban forms the authors identify.
The city has been hurting downtown vitality for years by tweaking the all-day rate to maximize revenue. The three Smart Parks near Pioneer Square fill up when they charge a competitive ($9 or $10) all-day. Shoppers and other short-term visitors then get turned away by the dozens between 11 am and 3 pm, and probably reconsider the whole idea of patronizing downtown stores and professional offices. A rate more like $12 leaves spaces available during the day, but those spaces bring in only a few hours’ revenue at the short-term rate.
Leaving the all-day rate high enough to make sure the garages never quite filled up would be more in line with the mission of Smart Park to provide short-term/visitor parking.
Tom raises a good point. You could actually get better parking outcomes for shoppers if you raised the all-day rate to the point where enough capacity was left over for them after the morning peak.
Or, simply sign a certain percentage of the stalls for short-term parking.
The lower floors of Smart Park are signed for short-term (4 hrs or less) parking but (ssh!) there isn’t any penalty or serious enforcement. Keeping 70% or more of the stalls short-term by signage would lead to some ridiculous behaviors, but might be a fun experiment to try–on the Planet of People Who Read and Obey Signs.
Here on Earth, I would again suggest regulating the number of people who drive into the garage intending to park all day by raising the rate to the appropriate level.
The city of Portland certainly enforces the on-street parking meters… why would it not enforce the 4-hour limit in the SmartParks?
Well, not to be rude, you could acknowledge that they’re not exactly the same situation and take it from there…
New York’s 2009 subway ridership decline is a result of lower gas prices and the recession. But until 2008, ridership increased. Between 1990 and 2008, subway ridership increased more than 50%, while VMT stagnated. Mode shares increased, too – the American Community Survey points to a slight increase in metro area mode share between 2000 and 2008.
The same trend cannot be observed for edgeless cities versus downtowns. Downtowns lost ground in the 1980s, gained in the 1990s, and apparently lost again in the 2000s. But even when they lost out in relative terms, they gained jobs in absolute terms, just more slowly than the suburbs.
Besides, edge cities can be converted to secondary downtowns, at great expense. Tysons Corner is planning such a conversion: the Washington Metro’s extending one line to serve it, and Tysons itself is planning to convert its street network to a grid. The Lang/DeFurgy paper discounts this possibility, but it’s happening, albeit sporadically.
Your claim in question is that “transit mode share in the [New York City] region is going up, not down.” You haven’t produced any evidence to support this claim. The ridership data I cited suggests that the claim is false. The recent drastic cuts in New York-area transit services casts further doubt on it.
The point is that the study you cited indicates a trend of job decentralization between 1980 and 1999. The Brookings study found a similar trend between 1998 and 2006.
EngineerScotty: I mean, first you would be requiring people to be sure whether they were staying four hours or less; then condemning people to drive the whole garage looking for an open long-term spot. If they didn’t find one they would have to make their way out of the garage, having wasted long minutes (ten or so). You would have to have a whole new system of enforcement which tagged cars (in the short-term spots) and then checked again four hours later to see who was still there, a lot like the enforcement of time-limited non-metered spaces around town. Then you would be giving people a ticket very possibly for just having put their car in the wrong spot. It’s a customer-service and branding nightmare. Compared to that, simply having a higher all-day rate is a proven method for pushing all-day parking down to a level that would keep spots open for short-term parkers.
“Commenter Pantheon dug into Canadian statistics and found that in the period 1996-2006, when Portland’s transit journey-to-work mode share was idling in the 12-15% range, Canadian cities posted these gains:
Ottawa up 2.2% to 19.4%
Montreal up 1.2% to 21.4%
Calgary up 3% to 15.6%
Vancouver up 2.2% to 16.5%
To be fair, these aren’t that much higher than Portland’s 12-15%, but it would be nice to think that Portland would show better outcomes than Calgary. ”
You’re comparing apples and oranges here Jarrett. Those Canadian numbers are metro numbers. That 12-15% for Portland is Portland city proper. Metro Portland is only around 6% or so IIRC.
And there is NO WAY you’ll ever see Portland even come close to Calgary’s share.
>>Vancouver now has a higher residential density than Manhattan.<< Nonsense. Vancouver City has about 13,000-14,000 per square mile. Downtown Vancouver has about 45,000 per square mile (about 100,000 in a little over 2 square miles). Manhattan has over 70,000 per square mile. I really hate it when people post lies. Is it that hard to look up numbers BEFORE you make a fool of yourself.
I don’t know if anyone wants to step in this river again, but: How do densities in Portland compare to those in Calgary, outside the downtown. Most of Portland–city and suburban–remains quite low density. I’m more familiar overall with Vancouver and am pretty sure that it’s outside of center neighborhoods are typically substantially denser than Portland.
How does roadway capacity, especially freeway capacity, compare in Portland and Canadian cities? Portland famously stopped one freeway and buried a (fairly short) segment of another, but there’s a substantial freeway network. To this Californian eye, freeway congestion in Portland is low.
Somebody brought up the two-way commute is good for transit theory. I have to say I’m skeptical, until transit has the dominant mode share. Before that, “reverse commutes” just easily degenerate into car commutes.
Jarrett, does this discussion have any bearing on your ideas (which I support) about frequent transit networks? I do know that Portland has taken an approach of identifying travel corridors before identifying modes.
Knew I forgot something! Like Jarrett, I don’t put much stock in those Brookings “distance to CBD” statistics for jobs. You pointed out that a farther away job could be on a transit line. Just as much, a job 3 miles from downtown could be in a difficult to reach area. Maybe this data would be useful for certain kinds of infrastructure that need to go in more area-wide like upgraded electrical service or sewer lines, but not for the geographically targeted needs of transit. I think the researchers used this data for job locations because it’s available and not much else is, but it’s not all that helpful.
The two-way commute is only really good for transit if the non-downtown employment concentration has disincentives to auto travel that are similar to downtown (e.g., parking limitations, parking fees etc.) and the transit system is very carefully set up.
In general transit systems are best suited to handle the “many to one” travel pattern — most areas can reach downtown reasonably easily by transit, whether bus, subway/LRT, or commuter rail, and because the downtown usually grows from the historic city centre, traffic service (road capacity, parking availability) is often more limited than the suburbs.
This is not usually the case for outlying employment concentrations, even for high-density employment nodes on a subway line. They are generally not as accessible to as much of the city/suburbs, and don’t usually have the same limitations on auto travel. (In fact, in some cities we even see peak-hour freeway travel becoming balanced between the so-called peak and off-peak directions as people commute by car to suburban jobs, in part because it’s too difficult to commute there by transit, and in part because there’s a perception that you’re traveling against rush hour traffic and therefore traffic shouldn’t be bad.)
I’m not libertarian, but I can’t figure out why cities are in parking at all, especially when they start building garages. Beverly Hills, CA, where I live, found itself on the losing end of a petition last year after it tried raising prices to pay for construction. The city wasn’t trying to profit, just balance its books. The petition cited the competitive advantages of surrounding mall businesses, all of which have private parking, and all of which (presumably) have parking costs built into their rents. Socialism at its finest.
If cities would get out of the parking business, they could also get out of parking regulation, the free market could do what it does best, and all those on-street parking spots could be turned into bike lanes.