Professor Becker Economics 145

Professor Becker

Economics 145

14 October 2010

The Effects of Transit-Oriented Development

A public transit stop, whether it is part of a heavy rail, light rail, bus, or bus rapid transit system, can have a very real impact on its surroundings. It can revitalize the neighborhood economy by providing new access points for retail, give locals without a car access to new job opportunities in other areas of town, it can reduce crime, it can change the racial makeup of a region. But much of the literature ignores these statistics, choosing instead to only focus on public transit’s effect on local property values. This is, after all, a reasonable research decision, as factors such as racial statistics would be anticipated to influence property values. However, by attempting to agglomerate all of these factors into one all-encompassing variable, it would seem that a critical part of the story is not being understood. This research uses a case study of the neighborhoods surrounding several Atlanta transit stations along the cities North/South MARTA line in order to create a more holistic vision of how these stops can shape an area and, in doing so, serve to widen the debate regarding Transit-Oriented Development.

Many, if not most, new transit systems in America are part of a broader attempt at Transit-Oriented Development (TOD). TOD is one of the most contentious topics in modern urban planning. The basic concept of TOD is that local governments can and should encourage decreased dependence on automobiles by creating convenient public transit nodes and supporting high-density development in the immediate vicinity of these points. This often comes in the form of new rail-transit investments and is lauded as an important tool in the fight against urban sprawl. New Urbanists see in TOD an opportunity to encourage high-density, mixed-use development like never before. More conservative purists, however, argue that these TOD neighborhoods represent unwanted and inefficient market distortions that cater only to middle and upper-class snobs too pretentious to ride a bus. The outcome of this debate will have critical implications on the built landscape of our urban environments for decades to come.

John Niles and Dick Nelson’s “Measuring the Success of Transit-Oriented Development: Retail Market Dynamics and Other Key Determinants” presents some of the basic determinants of success for TOD projects with a fairly pessimistic tone. The authors make sure to highlight the distinction between local success and regional success in TOD development. No TOD will realize its full potential unless it is part of a larger regional plan which incorporates numerous distinct TODs. Indeed this is one of the main limiting factors to the expansion of this model, as there are many communities that lack truly regional planning organizations and are therefore narrowly focused on neighborhood impact. The authors proceed to present other potential determinants which might enable success in TOD. Underlying many of these factors are a fundamental consumer interest in residential access to public transit.

Much of the literature in support of TOD is based on the almost utopist vision of the large—yet inevitably unquantifiable—externalities which may be derived from tightly knit walking communities, in which everyone waves at their neighbor on their nightly walk to the local grocery store. However, this meta-analysis focuses solely on the extent to which any new TOD is able to spark a shift away from automobile transit, which the authors consider to be one of the few truly measurable benchmarks of success in this field. Although well intentioned, this dismissal of any immeasurable externalities associated with TOD from consideration severely limits the conclusions of the authors.

Despite their restricted presentation of the consequences of TOD, the authors do provide an impressively comprehensive discussion of the factors which might lead to success in TOD. Street patterns, for example, are noted as one determinant of TOD success. This is one area in which New Urbanists have taken an immense interest. New Urbanists will often excitedly advocate for a return to a gridded neighborhood street system typical of many older suburbs built in the 1950s, in contrast to the more cul-de-sac, arterial design favored my many neighborhood developers and homebuyers in today’s market. They view this proposed shift as important for a number of reasons, not the least of which is that it would slow down automobile transit and encourage pedestrians to “take back the streets”. This gridded street pattern is therefore considered by some to be an essential part of TOD success, as it enables easier pedestrian access to transit nodes. The authors cast doubt on this assumption, citing earlier research questioning the true transit benefit of this type of design. However, it is difficult to believe that consistently gridded-streets, when combined with a more transit focused style of development, would not lead to marginal reductions in automobile use.

Joyce Chen et. al.’s “Transit Oriented Development and Cluster Developments” presents a much more positive spin on TOD, which the authors view as a wonderful tool available to modern urban planners. Like Niles and Nelson, Chen seeks to highlight the potential determinants of TOD success, providing Metropolitan Place in Renton, Washington as an example of a successful TOD. Chen, however, has a fundamentally different conception of what these factors are due to his more interventionist view of urban planning. She believes that the objective of transportation planners should be “surrounding a transit station with development appropriate to high-occupancy transit use” (p. 2) in order to encourage more public transit use. Chen advocates for short-term tax abatements, public incentives, and even zoning restrictions in pursuit of this goal.

Chen’s differing view of TOD is also informed by his conception of the barriers to this style of development. In addition to highlighting the absence of regional planning authorities as a major obstacle, Chen writes that “The biggest problem with this tool is that given a choice, people often choose socially and economically stratified neighborhoods.” (p. 4) The authors find little beyond classism and poor management to blame for unsuccessful transit-based developments and take it as a given that TOD is the preferable homeowner lifestyle. Yet if market demand truly has shifted towards TOD, it is difficult to imagine how this has not been echoed in increased private sector exploration of this concept. In presenting the relatively new idea of TOD, the authors provide a notably constricted argument in favor of this style.

Randal O’Toole’s “Defining Success: The Case against Rail Transit” argues that rail transit—be it cable cars, light rail, or subways—is an inefficient allocation of public funds in almost all scenarios. O’Toole details the massive rise in rail transit construction projects in the last few decades, blaming frequent construction overruns and ambitious ridership projections for veiling the true economic waste of these projects. The author also blames coercion by major real estate developers and the perverse incentives coming from the federal government that encourage these projects even in the face of net social welfare loss. The author provides detailed tables of operating costs, ridership, and overall profitability drawn from the National Transit Database to support his claims, and makes a fairly foolproof argument that buses are a more cost efficient alternative to expensive rail projects. The article also provides convincing evidence to refute the oft-cited environmental benefits of trains in comparison to other forms of public transit.

One argument often presented by rail-transit advocates is that the permanence of railway stops has the potential to spur high-density development in surrounding areas. The author of this piece suggests that this development is primarily due to localized government development incentives, which, when coupled with government subsidy of the railways, serves to create a government supported façade of positive economic growth. Yet it is difficult to imagine that these emerging TOD communities are springing up without some form of conceptual buy-in from developers and homebuyers. Although perhaps beyond the scope of this work, O’Toole fails to present a true economic cost-benefit analysis of rail-transit construction when viewed within the context of potential positive TOD externalities. Yet these externalities would need to be quite large to offset the inefficiencies which O’Toole identifies. Also, it is important to note that in sharply dismissing rail-transit and accompanying TOD neighborhoods, O’Toole is not eliminating the prospect of successful TOD communities which employ a public bus stop as their primary transit node.

Almost all of the hedonic study on the subject focuses on the spatial effect of transit stations on super-local real estate values. Although some may debate the true cause of much of this effect, most studies have found at least a small positive premium on housing that is relatively close to these stations (31, Pittsburgh). This effect is most pronounced in areas surrounding heavy rail stations, presumably due to their more permanent feel. However, even bus rapid transit has been shown to have a significant impact on area real estate, with a property value located 1000 feet away from a station being valued $9,745 less than a property located 100 feet away (Pittsburgh).

In contrast to many studies on the matter, Bollinger and Ihlanfeldt’s “The Impact of Rapid Rail Transit on Economic Development: The Case of Altanta’s MARTA” investigated how rail transit has the potential to shape a community beyond property values. This study of the MARTA system sought to correlate population and employment statistics in a given census tract with a variable which represented the percentage of that tract which fell within 0.25 miles of a MARTA station. The earlier “Transit Station Area Development Study” (TSADS) had divided the Atlanta stations into five categories: high-intensity urban node, mixed-use regional node, commuter station, community center, neighborhood station. Bollinger and Ihlanfeldt found no spatial correlation with total population. The study also found little impact on total employment in the station areas. Although there was a higher rate of public sector employment in stations identified as high-intensity urban nodes or mixed-use regional nodes, this increase was offset by a similar decrease in private sector employment. The authors use these findings to bolster their initial claim that rail-transit is an inefficient use of public funds.

Bollinger and Ihlanfeldt’s critique of rail spending is limited by its use of census tract data. Although the authors note that this study, conducted in 1997, was the first of its kind to use data as narrow as the census tract, it would seem that these areas would still be too large to show symptoms of any significance from a single rail transit station. The studies spatial solutions, while a valiant effort to accommodate overly broad data, remain unsatisfying. In addition, it seems unlikely that the full impact of these stations, constructed as late as 1993, would truly have shown through in a paper published just 4 years later. The study does, however, provide a valuable starting point for this research.

Like many cities in the Southern United States, Atlanta, Georgia is a fairly auto-centric metropolis. This can be seen largely as an effect of Atlanta’s relatively recent emergence, with much of its growth beginning in the 1970s and continuing through to this day. This late growth came after the automobile had been firmly established as the primary mode of transportation. This also came during a time of great suburbanization and the so-called “white-flight” of American cities. This late growth, along with the lack of any major geographic boundary, has contributed to Atlanta’s status as one of the most sprawling metropolises in the nation.

The United States as a whole has fundamentally shifted its urban landscape since the 1950s. In the time after WWII, the nation experienced a tremendous economic boom. Fueled by unrivaled economic prosperity, broad social shifts, and supportive federal policy, Americans began to move away from the center of many major cities. Beginning with Levittown in the late 1940s-early 1950s, many families and their new baby-boom children began to move to the suburbs.

It is this shift that many are now blaming for a number of present day American issues. Some blame sprawl for environmental degradation because of how it expands city footprints and fosters auto-dependence (Johnson). Others connect sprawl with a decline in physical activity and rise in obesity (Ewing, Relationship Between Urban Sprawl and Physical Activity, etc.). Still others have investigated the potential association between urban sprawl and levels of income inequality (Wheeler, St. Louis Fed). The extensive literature attempts to pin every societal problem of the last half century on the expanding American city.

There has, however, been a backlash against urban sprawl that has emerged recently. Massive efforts are underway to revitalize many American central business districts and reel people back into town from distant suburbs. A critical piece of this effort will be public transportation, which the Obama administration has vocally supported.

Atlanta was relatively early to create a mass transit system, with the first MARTA train beginning operation on June 30, 1979 (MARTA website). Station by station, the system grew outwards from its initial downtown core. The system is designed with two main lines bisecting each other, with one running on a North/South axis and the other on an East/West axis. Average ridership for the system was estimated at 482,500 passengers per day in 2009 (Dawson, TRP Report). Although MARTA ridership fell by 3.60% in the second quarter of 2010, this service is still reasonably popular amongst locals (Elliot, BizJournal). Recently, the MARTA network has grown to include bus rapid transit options. Atlanta officials are even in the process of planning a controversial tourist-friendly streetcar service which would be funded largely by a recent $72 million Department of Transportation grant (Faussett, LA Times).

One of the reasons that Atlanta was selected for this case study was that the Atlanta government uniquely did little—beyond rezoning—to actively spur investment around its MARTA stations (Marta Bollinger), allowing the outside effects of these stations to be more easily isolated. This is mostly due to the paltry funding that MARTA has received from both local and state governments. Despite this lassiez-faire approach, many of the new developments in the area are anecdotally positioned in the near vicinity of the rail arteries (Elliot).

Another factor that makes Atlanta interesting is the racial implications of MARTA. Atlanta was a critical battleground in the Civil Rights Movement and has a significant African-American population. MARTA, which is oftentimes sarcastically referred to as “Moving Africans Right Through Atlanta”, has a ridership that is largely African-American. Some have argued that the low funding for MARTA is rooted in the racism of the whiter suburban counties in the region. It is easy to see how MARTA might fit into the more local debate surrounding Durham, North Carolina’s DATA service, which is seen as largely catering to local minorities. Indeed most public transportation systems in the United States do seem to serve a large minority population. Although this has the potential to introduce unwanted confounding factors into this analysis, it is an extremely interesting case study in the potential effects of public transit.

In order to gain a more comprehensive understanding of how Atlanta neighborhoods changed as rail transit stations were built, a number of these neighborhoods were examined in depth. Bollinger’s study was unique in that it was one of the first of its kind to look at data as narrow as a census tract. I chose to take this analysis one step farther and examine the census block in which these neighborhoods were built, as this is the smallest division for which I had access. These small blocks will presumably be precise enough to capture the walking distance surrounding the station point.

In contrast with Bollinger’s study, I examined time-series data in my analysis. The motivation for this was that I wanted to gain an understanding of how these neighborhoods changed over time. By observing a more long-term picture, I hope to advance the understanding of both short-term and long-term effects on station areas. Bollinger’s study of the MARTA system sought to correlate core demographic statistics with a variable which represented the percentage of each Atlanta census tract which fell within 0.25 miles of a station. I chose instead to compare data from the station areas with aggregated data from both the central city of Atlanta and the broader Atlanta metropolitan statistical area.

I used the SimplyMap program for the bulk of my analysis. Because the SimplyMap database does not contain data from before the 1980’s census, I chose to limit my station observations to those built between 1980 and 1990. Luckily, Bollinger’s analysis provides a comprehensive listing of all station opening dates in the MARTA system (FIGURE). The 12 stations selected include North Avenue, Midtown, Chamblee, Civic Center, Brookhaven, Lenox, Garnett, West End, Oakland City, East Point, College Park, and Arts Center. These stations, which are all on the North/South MARTA line, constitute a diverse set of neighborhoods that will hopefully reduce selection bias. Indeed, this sample includes stations in all five categories (commuter, community center, mixed use, high intensity, and neighborhood) defined in the TSADS report. Although this is a reasonably small sample from which to draw and there is some danger of selection bias, this study is intended as a case study and is not attempting to demonstrate any quantifiable causality.

Using SimplyMap statistics of block-groups which contained the desired stations—whose addresses were scraped from the MARTA website—I was able to obtain detailed data for every area. This data was then exported to an Excel spreadsheet, where I compared it with data for the central city of Atlanta as well as the broader Atlanta MSA which was taken from the State of the Cities Data Systems which is provided by the United States Department of Housing and Urban Development. Using basic statistical analysis, I then calculated two-sided t-scores for composite averages of each of the variables, which include household income, percentage of housing units which are renter-occupied, percentage of housing units which are owner-occupied, and racial statistics for whites, blacks, and Hispanics.

Then I proceeded to extract data and calculate t-scores for all censuses that have been taken since the stations were built. In this way, I hope to longitudinally track how these neighborhoods are doing decades after their stations are built. A major critique of many spatial econometric studies on the effect of transit stations is that, in examining only one time set of data on stations that were built somewhat recently, they fail to capture the longer term effects of stations. With data that extends across decades, I hope to examine what happened in these Atlanta neighborhoods after the usual public/private development spike once the real estate market has had adequate time to account for the new station.

As mentioned earlier, issues of race are intimately tied to MARTA’s development. The unofficial acronym for many remains “Moving Africans Right Through Atlanta”, and MARTA ridership is predominately African-American. Yet beyond simple ridership surveys, there has been little effort to quantify the differences in development between Atlanta neighborhoods with and without the stations. Presumably, if African-Americans are in any way more partial towards MARTA—and rail transit in general—then one would expect to see a spike in African-Americans living in the station areas that is disproportionate to the city or region as a whole. One might argue that racial statistics are too tied to socioeconomic patterns to derive any meaningful insight.

Perhaps, the argument goes, rail transit appeals to poor people, who just happen to be disproportionately African-American. It is certainly true those low-income residents without access to cars or without the money to pay for gas benefit greatly from public transit. These poor residents benefit from direct savings from using public transit instead of individual transit, expanded access to job opportunites farther from home, and likely increased patronage of local retail outlets which tend to be disproportionately staffed with low-skilled workers. While the magnitudes—and even signs—of these and other factors are certainly up for debate, it is difficult to imagine a framework in which the poor would be uniquely hurt by investment in a nearby public transit node.

Yet we will later see that, at the same time these census blocks were becoming disproportionately African-American, they were also reaping significant strides in household income beyond that which could be expected of a typical Atlanta census block. When combined with the overwhelmingly African-American ridership statistics, these data begin to make a compelling case that there was—and perhaps is—something about MARTA which has made it tremendously popular among the African-American community.

References

Bollinger, Christopher R., and Keith R. Ihlanfeldt. "The Impact of Rapid Rail Transit on Economic Development: The Case of Atlanta’s MARTA." Journal of Urban Economics 42 (1997): 179-204. Web. 6 Dec. 2010.

Chen, Joyce, Mark Hamilton, Nick Kindel, Ian Macek, and Meghan Pinch. "Transit Oriented Development and Cluster Developments." 1-11. Web. 23 Sept. 2010. .

Dawson, Christie R. Transit Ridership Report: Third Quarter 2009. Rep. American Public Transportation Association. Print.

Elliott, Mark. "MARTA Rail and Offices." Atlanta Business Chronicle 20 Oct. 2010. BizJournals. 20 Oct. 2010. Web. 6 Dec. 2010.

Grass, Gail. "The Estimation of Residential Property Values Around Transit Station Sites in Washington, D.C." Journal of Economics and Finance 16.2 (1992): 139-46. SpringerLink. Web. 6 Dec. 2010.

Ihlanfeldt, Keith R. "Rail Transit and Neighborhood Crime: The Case of Atlanta, Georgia." Southern Economic Journal 70.2 (2003): 273-94. ProQuest. Web. 6 Dec. 2010.

MARTA Homepage. 2009. Web. 6 Dec. 2010. .

Niles, John, and Dick Nelson. "Measuring the Success of Transit-Oriented Development: Retail Market Dynamics and Other Key Determinants." Proc. of American Planning Association National Planning Conference. 31 Aug. 2006. Web. 23 Sept. 2010. .

.

Perk, Victoria A., and Martin Catala. Land Use Impacts of Bus Rapid Transit: Effects of BRT Station Proximity on Property Values along the Pittsburgh Martin Luther King, Jr. East Busway. Rep. no. FTA-FL-26-7109.2009.6. US Department of Transportation, Federal Transit Administration, Dec. 2009. Web. 6 Dec. 2010.

Directory: urbaneconomics -> files -> 2010
2010 -> Jiawen Cheong Economics 145 Literature Review

Download 34.78 Kb.

Share with your friends: