Avoiding the Unintended Consequences of Transit-Oriented Development

We see it over and over again in our cities. Migration out of central cities hollows out neighborhoods and leaves the people who remain struggling with the consequences of disinvestment. But when development returns to urban areas, the arrival of new residents can impose burdens on people who never left. Often, as amenities come into an area and crime goes down, property values rise and poorer residents can no longer afford to live there.

The addition of light rail has been linked to higher rates of car ownership, as compared to the Metropolitan Statistical Area as a whole, but that doesn’t mean we should stop building light rail. Image: Dukakis Center (##http://www.dukakiscenter.org/storage/TRNEquityFull.pdf##PDF##)

Even when the new development is built around transit, which can lower transportation costs for low-income residents, unintended consequences can ensue.

Researchers with the Dukakis Center for Urban and Regional Policy have recently reached some provocative conclusions from their study of gentrification and transit-oriented development. Without proper planning, they found, new development near transit can lead to stratified neighborhoods and higher rates of car ownership.

They also offered some solutions to ensure that transit-oriented development achieves its intended goals. The solutions include preserving affordable housing and restricting parking in new developments.

Historically, the authors note, transit-rich neighborhoods tend to be diverse. The low-income people and people of color who live there often don’t have cars and they depend on public transportation. They also usually rent their homes — and since rental housing turns over faster than owner-occupied homes, this speeds along the process of gentrification when new transit options come to a neighborhood.

Rents go up as transit arrives (often along with new shops and restaurants) and more affluent people move in. And guess what? Those wealthier people tend to have more cars. That’s the fundamental paradox: the people who are attracted to transit-rich neighborhoods – and have the money to pay more to live there – don’t use transit as much as less affluent people who can get priced out.

The authors stop short of calling this pattern “displacement” – they point to “normal processes of housing turnover and succession.” But what’s clear is that the people moving in are from a different income demographic than those moving out (though the researchers say the racial makeup tends to stay the same).

Income-based housing stratification and more cars are not the outcomes planners want from transit or transit-oriented development. The challenge is to keep development around transit from becoming too exclusive and too car-oriented. How can communities do this?

The report comes with an entire policy toolkit for communities planning new transit (especially light rail, which, according to the research, brings about even more profound change than other forms of transportation.)

In San Leandro, California, they’ve implemented a comprehensive strategy to preserve existing affordable housing and add more. In Minneapolis, the Longfellow neighborhood negotiated benefits that the city then incorporated into the development approvals — making the deal binding. In all cases, ensuring broad public participation in the process was essential.

But having your say is one thing. Actually being able to afford your own home next year and the year after — that’s what counts. In Denver, Charlotte, and the Bay Area, communities used transit-oriented development acquisition funds to buy or preserve affordable housing before transit projects came in and drove up land prices and property values.

Some other tools in the equitable TOD toolbox:

  • Housing trust funds, which are dedicated sources of public funding for affordable housing
  • Low-income tax credits, allocated by state housing agencies to developers to provide money for affordable housing
  • Tax increment financing districts, which use the revenue from the higher property taxes in the surrounding area to help finance the building and preservation of affordable housing
  • Inclusionary zoning ordinances requiring some proportion of new units to be affordable (usually 10-25 percent, but sometimes more)
  • Housing incentive programs that fund transportation-related livability infrastructure in affordable housing projects, which reward local communities for the creation of affordable housing near transit

These are ways to maintain affordability and to keep the existing residents from having to leave. But you also have to incentivize transit use among the new residents, who have the means to drive a lot if they choose. Stephanie Pollack, the lead study author, says it’s just as important to have “transit-oriented neighbors” as to have “transit-oriented development.” One of the most important levers, it turns out, is parking policy.

Here’s what she told an audience at Rail~volution about one transportation package she negotiated in Boston.

This is what the package looked like: it was designed to get at the issue of creating transit-oriented neighbors who would use the transit. One parking space per unit, priced separately from the condo… shared parking spaces in the same garage so people would make their second car a shared car; and a free annual transit pass for your first year after purchase, provided by the developer, who by the way will spend way more on a second sub-surface parking spot for each unit than that annual transit pass will cost them.

You have to think about the market housing in transit-oriented neighborhoods. We’re so focused on ‘we need affordable housing’ – because we do – but we still want the people living in the non-affordable housing to be good transit neighbors.

“Unbundling” parking from the price of the unit is key. People don’t notice the cost of car ownership when it’s folded into the cost of their housing, and car-free residents effectively end up sharing the cost of providing parking for car owners — they don’t get the full financial pay-off of eschewing a car. But when faced with the prospect of paying $100 a month for parking in your own building, the benefits of going car-free – especially in a neighborhood newly equipped with good public transportation – suddenly sound like a good deal compared to the costs of owning a car.

24 thoughts on Avoiding the Unintended Consequences of Transit-Oriented Development

  1. Judging from this summary, they seem to miss the most obvious solution. If building new transit drives up property values around stops so drastically, that is a sign that we do not have enough locations that are convenient to transit: the price goes up because the demand for transit-oriented development is greater than the supply. In the long run, the best solution is to keep building more transit and more transit-oriented development until supply catches up with demand.

  2. It’s important to create disincentives to car ownership, but more important is making sure public transit remains reliable and attractive. There are already plenty of reasons not to drive in a dense city — bad traffic, difficulty parking, having to monitor your alcohol intake, lack of exercise. But if the trains don’t run on time, cars will remain attractive.

    I’m one of those gentrifiers who moved into a developing part of SF and brought my car with me. I love public transit and would use it all the time, but just last night I was 30 minutes late to a friend’s house when a promised Muni bus didn’t show up. You can disincentivize cars all day long, but most people won’t tolerate persistent frustrations and faux pas in the name of transit advocacy.

  3. I like that our planners continue to expand their horizon. But it needs to keep expanding. Transit and housing and cars and jobs and gentrification and schools and design and architecture and planning and public space and property values and race are all interconnected. Integrated solutions are what will work. And I’m glad were’ moving in that direction. Car sharing is an essential ingredient that is still missing. Public transit is not up to snuff yet, and is nowhere near good enough to make people feel comfortable giving up their personal automobile. Yet, an affordable car sharing program that is done on a regional basis, that promotes community, is something that can fill the void.

  4. Unbundling is definitely a good start. I am all for the choice to drive, if people are willing to pay the real cost to do so.

    Charles Siegel has a very important point that I’d like to add to. Not only can we lower the cost of living in cities by building more urban housing, we can also lower the cost of living in our biggest cities by re-developing our smaller regional cities. The cost of living in our biggest cities is so high because the smaller cities, for the most part, have no urban draw anymore.

  5. Raise the price of street parking. A lot.

    The balance of Matt’s issue of the value of having the car vs. the value of the easy transit neighborhood would probably be changed if it cost him $10/hour for street parking, day and night. He’s wealthier than many of his neighbors, but he’s probably not that wealthy. If it cost that much to park on the street, parking lots, including the one possibly in his building, would raise their rates because they could. Even if Matt has a free parking spot, the extra income he might get from renting out his spot for $1000/month might tempt him to sell the car.

    If the rate is so high that much of the street parking doesn’t get used, I’m sure we can think of other uses for that space.

  6. A lot of people use both transit and cars. Transit during the week to go to work, cars to go out to the suburbs to visit family, or to go shopping in the malls/Ikea. So car owners in the outer boroughs are not necessarily using their cars during peak hours and adding to congestion.

  7. Charles, let me be the first to say it–Transit Sprawl!

    I looked at the report and what they aim for isn’t only transit-oriented development but also transit-oriented neighbors–paying riders. So they want to discourage commuting by car by limiting car parking. OK, but why not eliminate bike racks as well? Why can’t we just ‘bundle’ residency with transit-usage(“Your house keys also work at the turnstile. All you pay is the monthly rent!”).

  8. A few observations:

    * It is well known that vehicle ownership and use increase with increased income, and transit use generally decreases with increased income. To a large extent, the results reflect those well-known relationships. The study did not control for the wealth effect. It would be worthwhile to know if wealthy households in transit neighborhoods used transit at higher rates than comparably wealthy households outside of transit neighborhoods.

    * Because the sample size was small in many cases, the researchers did a re-analysis of neighborhoods that had more statistical significance. The results for the rate of change of transit use for commuting relative to the metro area were: 31% more; 50% equal; 19% less.

    * The actual transit mode share is much more relevant than the relative rate of change. But the study only analyzed the latter. Neighborhoods with a relatively low rate of change may already have a high transit mode share — and based on the the list of neighborhoods studied, that does seems likely.

    * The study did not consider transit uses other than commuting. About half of all transit trips in the US are for purposes other than commuting.

    * The study confirmed that transit can be remarkably successful at attracting new development and a wealthier clientele. The full report included a number of policy actions for planning TOD, increasing transit use, and ensuring affordable housing. However, design factors that encourage residents to walk or bike to transit were ignored, even though they can have a large impact.

  9. I agree with others who have posted here that:

    (1) Rent goes up because there are not enough transit-rich neighborhoods to meet the demand–so create more of them!

    (2) An increase in middle-class and affluent transit riders is a very good thing, because they are in a far more favorable position to demand and influence transit maintenance, upgrading, and expansion than the poor.

    (3) As transit is maintained, upgraded, and expanded, it attracts more riders and more transit-oriented development, which helps ease the higher housing costs mentioned in #1 by increasing supply.

  10. The “TOD” projects here at rail stations in Los Angeles produce an overabundance of car parking, loading docks, truck access, etc. These same projects lack public restrooms, bike storage, well lit corridors, an architecture that welcomes pedestrians, traffic calming, and safe crossing points.

    So, in the end, you get a bunch of rich dickheads driving in and out of their new “urban” condos while the rest of us are in a system meant for notional human beings (the kind that get rendered in architectural sketches), with no amenity for the real thing.

  11. “The addition of light rail has been linked to higher rates of car ownership, as compared to the Metropolitan Statistical Area as a whole, but that doesn’t mean we should stop building light rail.”


    Light rail projects are massively expensive. Most of the new riders they attract are wealthy people who already have other options. To me, this is a reason to stop building light rail.

    Call me Herbie “So tired of the streetsblog rail fantasy” Huff

  12. At a local scale, transit-oriented gentrification can bring more cars into a given area. The question is where would the gentrifiers live if they didn’t live in a TOD? They wouldn’t just disappear. They’d live in other, less transit-oriented places, bringing at least as many (if not more) cars to them. To really understand what’s going on, you’ve got to look at a regional scale.

    More locations than rail stations are definitely needed. For all its lapses, Muni in San Francisco provides a service over many parts of the city that people with options are willing to use as primary transit. The phrase that’s often heard is that “San Francisco is one big TOD,” even though much of the T is, gasp, buses. Those conditions need to be created in other cities, through efforts like Portland’s frequent service bus lines.

  13. Okay, one thing about this analysis: is the only measure of a TOD going to be based on the state of automobile travel?

    We had a big bullshit LA Times article about how our latest subway extension wouldn’t improve car travel in the city. Big whoop. It isn’t about improving car travel – it’s about improving (in the case of the subway) train travel and public transportation.

    With TOD, there are a host of other more important questions to be asked.

  14. I do think one of the main purposes of transit should be to move the poor.

    Except in New York City, where there is very little free parking and an extensive subway system, most transit riders are poor.

    This is not my opinion, this is a behavioral fact. Most people with other options choose to drive for most trips.

    Obviously, as someone who chooses not to drive for many trips myself and who believes there are immense negative consequences to a car-centric transportation system, I believe it is important that we provide other options. But I don’t think rail is the way to go, because it is so expensive.

    I think rail is charity for wealthy people who do not need it. Those who are really committed to other forms of transportation (i.e. those who do not own cars) would be better served, in my opinion, by systemwide improvements. These include cheap fares, networks of bus-only lanes, well designed and maintained stations and stops, and very short headways.

    All this is not nearly as sexy or political as rail. It does not lend itself to ribbon cutting. But I think it is a better use of our money. Reasonable people can disagree, but few and far between are the commenters on Streetsblog who are willing to even consider the idea that rail might not be the best use of our money – @Wanderer being an exception 🙂

    I’m sorry if I came off as insulting, it just gets tiring to hear Streetsblog readers incessantly cheer for rail, seemingly without concern for how rail projects have actually performed over the past 20 years in terms of getting more people to ride transit and give up their cars, and seemingly without concern for rail’s enormous costs.

    This article is about how car ownership goes UP in TODs! And no one on Streetsblog has a real eyebrow to raise about that? Or any comments on how the presence of rail seems to be doing very little to influence mode choice, and how more important factors seem to be car ownership and parking? @Capn Transit, you asked about the goals of transit. I think many people think one of the goals of transit is to get people out of their cars, but we need to get real and recognize that transit very very rarely accomplishes this. That’s what this article is about.

    It seems we as a community have stopped valuing logic and have become rail cheerleaders at any cost. We will twist the argument whichever way it needs to be twisted in order to reach the conclusion that we should spend more money on rail projects.

    I speak both about the responses to this post and about the general subway cheerleading that takes place on LA streetsblog.

  15. I think many people think one of the goals of transit is to get people out of their cars, but we need to get real and recognize that transit very very rarely accomplishes this. That’s what this article is about.

    No Herbie, that is not what the study concludes. The study is not about actual transit mode share. The study is about rates of change of transit use in TODs relative to their metro area. Neighborhoods with relatively low rates of change can still have high transit mode share.

    Furthermore, when the statistically significant TODs were re-analyzed, the great majority (81%) had rates of change that were equal to or greater than the metro area.

  16. @Laurence Aurbach – The study observes higher rates of car ownership in light rail TODs. I believe that is the take away point from the Figure above.

    If you want to get technical about it, the study’s authors say that the study is about how “transit investment can sometimes lead to undesirable forms of neighborhood change.”

    @Charles Siegel – what you say – and what Streetsblog echoes in the featured quote – I don’t believe it. Streetsblog should be embarrassed that they willingly echoed such poor economic logic. Living in a TOD is expensive because of a lack of supply? And high demand? You forget a key variable – costs. Look, large numbers of BMWs are supplied all over the country. No matter how high the supply, that will never make BMWs cheap. The same is true with TODs.

    Living in a TOD is expensive because building rail transit is very expensive. Building a bigger “supply” of rail won’t make the costs of rail go down. Living next to rail will always be a rare luxury, unless we break all of our government budgets in a really unprecedented way.

  17. No Herbie, the figure above does not show higher rates of car ownership in light rail TODs. The figure above shows higher relative rates of change of car ownership in light rail TODs. There is a big difference. The relative rate of change of car ownership can be high while at the same time the actual rate of car ownership can be low.

  18. @Laurence

    I’m confused. The figure shows that the percentage of households with no cars in light rail TODs is almost 20% lower than the percentage of households with no cars in the MSA as a whole.


    What do you mean by “the actual rate of car ownership can be low”? Low relative to what?

    You claim that “there is a big difference” between relative rates and absolute rates, but when we are talking about a social phenomenon like car ownership, there are really no absolute benchmarks. We can compare car ownership in a particular place to the rest of the city, or the rest of the nation, or etc. In this case the authors compared it to the rest of the city, and found that there was a higher percentage of car-free households in the city as a whole than in the light-rail TODs.

    Anyway, I don’t want to get bogged down in these details, that’s not really my purpose here. My purpose is to raise the point that no one on streetsblog seems to take the concerns raised by this study seriously. Few people on streetsblog consider rail construction to be a transfer of wealth from the general populace to the (generally affluent) new riders that rail transit and expensive TOD developments attract. Even the authors are not willing to fundamentally question whether light rail development is the best idea; they just offer a toolkit of ways to ameliorate the gentrification that they observe.

    I think transit advocates and people who care about alternative transportation need to be taking findings like this very seriously. And I put forward that perhaps the answer is to stop making massive investments in rail and to redirect the money toward operational improvements that will actually help the car-free and car-lite households. Those same households, as the study hints, are spread throughout the MSA and not necessarily located near a rail stop.

    Anyway, I feel like I’m repeating myself here with little success in getting a substantive reply. I don’t really want to nitpick about the figure. I want to talk about the big picture in how transit money gets spent and who those investments benefit.

  19. Herbie, the methodology of the study is outlined on pages 21-22 and Appendix A at the end. The study collected census statistics from 1990 and 2000. The study compared the change that happened between 1990 and 2000. The study reported “differences in the rate of change between each station area and the surrounding metropolitan area.”

    Therefore, the figure does not show that the percentage of households with no cars in light rail TODs is almost 20% lower than the percentage of households with no cars in the MSA as a whole.

    The figure does show that almost 20% of households in light rail TODs had a percentage change from 1990-2000 that was less than the metropolitan area.

    I agree with you that the way the numbers are reported is confusing. The authors provide some but not all of their database, and their explanation of their methodology is confusing. So my interpretation could be wrong, but it is my best effort at understanding the numbers.

    To give an example, Denver has a light rail system. The authors report the percentage of households with zero cars in Denver station areas in 2000 ranges from 30%-43%. That’s a drop from the 1990 percentages. But it’s still much higher than the City of Denver’s percentage (<a href="http://www.bikesatwork.com/carfree/census-lookup.php?state_select=ALL_STATES&sort_num=5&lower_pop=250000&upper_pop=999999999&first_row=25&show_rows=25"which is 14%) and it surely must be higher than the metropolitan area overall.

    As I said above, I don’t think relative rate of change is nearly as relevant as actual percentage and actual mode share.

    On a different topic, the cost of housing and commercial space in TODs has nothing to do with the cost of building rail transit. TOD developers and residents do not directly pay the cost of building rail transit. The cost of building rail transit is funded by taxes and fees shared across an entire city, county, and state, as well as federal contributions.

    The cost of housing and commercial space in TODs is highly related to the cost of land, as well as market demand. Those factors wouldn’t raise end-user costs so much if the supply of TODs was increased.

  20. @Laurence. Agreed. Confusing.

    Don’t you think that the costs of TODs have to do with the rarity of rail transit stations? Might the explanation for the observed gentrification have to do with the rarity of rail transit stops and the intense politicization of their siting? It makes sense to me that cities would try to use their few rail stops to generate the maximum amount of tax revenue by encouraging high-income developments.

    Yes, you could build affordable housing near rail, IF the price of land is low enough and you build cheaply. Will cities facilitate affordable housing development near rail transit when they have made major investments in the adjacent rail line? I don’t think so. Rather, I think cities have tried to get the maximum return on their rail investment by aiming these developments at the wealthy, and I think that will continue to happen. Furthermore, will the cost of land really be low? If rail stations are rare and in high demand, this will drive the cost of prime TOD land up. Given the exorbitant costs of building rail, I don’t see the rarity of rail stations changing anytime soon in any city.

    Does that make sense? I hope my economic point is clear.

    So I believe that the costs of TOD are inextricably linked to the very high costs of rail. Thus, I believe that the vast majority of TODs will continue to be expensive to live in and will attract people who own and drive cars regularly. I agree with the study authors that we should be limiting parking in TODs in order to both decrease the cost of the housing and also to attract folks who deserve to live near excellent transit, i.e. the folks who are committed to not driving.

  21. The description of the figure that I gave in my post #21 may be wrong. Here is a correction:

    The figure shows that for all households in light rail station areas, the rate of change is nearly 20% lower than the metro area overall.

    I definitely agree there are political actors who are strongly interested in limiting the number of transit stops and buildable land near transit. Landowners in existing station areas would like to limit the competition, auto interests want the money to be spent on roadways and parking, and NIIMBYs want to block new development for fear that their quality of life will be worsened.

    But I disagree that transit always has to be exorbitant in cost. This article by Ed Tennyson says that light rail at street level costs $30 million per mile, while a 6-lane freeway that carries the same number of people per hour costs $102 million per mile. In some cases rapid bus is more suitable and cheaper, and in some cases light rail is better, as well as cheaper over the long run. Todd Litman’s “Evaluating Public Transit Benefits and Costs” is a good guide to the issues, and he provides a number of related reports.

    The Mixed-Income Transit Oriented Development Action Guide is a suite of tools and strategies for promoting affordable housing in station areas. Reconnecting America provides several reports that cover similar topics of affordability in station areas and and financing of transit. Many of these recommendations are based on best practices that have already been instituted by cities, counties and regions, showing it’s possible to overcome the political forces that restrict transit. However, you’re certainly right that there is much more work that needs to be done to reduce the cost of building transit systems and the cost of developing near transit.

Leave a Reply

Your email address will not be published. Required fields are marked *


Brookings: Transit Access to Jobs Is the Missing Link

If you’re a middle-income person living in the Philadelphia metro area, there’s an 85 percent chance you live within three-quarters of a mile of a transit stop, and you probably have to wait about 12 minutes for a bus or train. But if you’re looking for work, beware: only 20 percent of the jobs in the […]

Can Transit-Oriented Development Lift All Boats?

Streetsblog San Francisco reported earlier this week that the Metropolitan Transportation Commission has made a $10 million funding commitment to a mixed-use affordable housing project in the Tenderloin neighborhood, a convenient two-block walk from the nearest Muni stop: The development at 168 Eddy Street would provide 153 new apartments reserved for low-income families and space […]

Can Transit Expansion Produce Sprawl Like Highways Do?

Here in the Washington, D.C. area, our Metro system is expanding. A new Silver Line will go all the way to Dulles airport and beyond, into exurban Loudon County. The projected station stops are named for highways, not neighborhoods or landmarks: Reston Parkway, Route 28, Route 606, Route 772. Ten of the 11 new stations […]

Can Phoenix Reinvent Itself as a Transit City?

Perhaps no other city in the country has the reputation for sprawl that Phoenix does, and it is well deserved. This is a city built around the car — until 2008, sprawling suburban housing in Maricopa County was the driving force of the regional economy. Phoenix got a rude awakening in 2008, when the housing […]