Crime and Transit: What’s the Real Story?
This is a guest post written by Daniel Yi, an intern with the Central Maryland Transportation Alliance during the summer of 2018. He attends River Hill High School in Howard County.
I had the opportunity and privilege of interning at the Central Maryland Transportation Alliance during the summer alongside Brian O’Malley, the President and CEO, and Eric Norton, the Director of Policy and Programs. There, I was assigned a project in which I would compare whether criminal activity was more prevalent around areas near transit compared to areas not near transit, specifically the Maryland Transit Administration’s (MTA) Light Rail stations. Using that data, the Transportation Alliance would determine if crime was influenced by the presence of transit stations or if there was no significant connection between the two.
The study focused on three jurisdictions, Anne Arundel County, Baltimore County, and Baltimore City and analyzed the occurrence of Part 1 crimes (murder, rape, theft, robbery, arson, grand theft auto, assault, etc.). When I first embarked on this research project in the beginning of July, this topic was relatively quiet. However, after the southern portion of the Light Rail in Anne Arundel County was shut down for flood maintenance in late July-early August, some residents urged their representatives to shut it down completely due to fears of increased crime and deteriorating quality of life in the neighborhoods around the station. In response, Anne Arundel County Executive Steve Schuh sent to the Maryland Secretary of Transportation a letter that called for two Light Rail stations to close and service to be reduced. As a result of these developments, this research began to grow and is designed to determine whether perceptions of increased crime are justified.
In order to establish a viable comparison between areas near transit and areas not near transit, the methodology involved plotting the addresses/coordinates for the stations on a Geographic Information System (GIS) map, creating a half-mile radius around each station, and looking at the crimes committed in that space. Then, in order to determine whether crime is more prevalent in areas near transit, we chose control locations that were outside of the half-mile radius of the stations. Then, we followed the same process; using the control locations, we created a half-mile radius and analyzed the crimes within that area. Since the Transportation Alliance did not have direct access to a criminal database, we contacted the respective police departments in each jurisdiction, with the first, and arguably the most urgent area being Anne Arundel County.
Prior to contacting the Anne Arundel County Police Department, I searched through public/open records online. One of the more useful sites to analyze criminal trends and data was SpotCrime, which is headquartered in Baltimore City. I contacted the organization, asking for more information and access to their records, and was referred to a representative of the Anne Arundel County Police Department. I sent him the list of station and control addresses for him to collect data on. These locations included the Cromwell, Ferndale, Linthicum, North Linthicum, and Nursery Road Light Rail stations and were compared against various locations in the county, consisting of commercial and residential properties not near Light Rail stations. The department sent me reports on all the crimes that took place in 2014–2017 around a half-mile radius of those locations and included an aggregate report of northern Anne Arundel County, which had all the locations within its borders. After receiving the data, I averaged out the Part 1 crimes and found a very interesting result:
As seen in the bar graph above, the averages for each category were added and visualized in a bar graph. Now, keep in mind that the non-transit addresses were nowhere near the actual light rails, so those modes could not influence activity in those locations. What seems plausible from looking at this graph is that criminal activity is largely influenced by the types of services offered in an area and how the land is used. For example, when comparing the Cromwell Station against Arundel Mills Mall, locations that are approximately five miles from each other, there is a significantly higher number of thefts and robberies in the latter than the former. This trend could be explained by the simple fact that the presence of Arundel Mills Mall leads to a higher volume of people, increasing the opportunities to rob individuals and shoplift from stores. Unlike malls, light rail stations, in and of themselves, do not really offer anything valuable. Crime rates tend to be higher where there are concentrations of people, cash, and/or merchandise, regardless of whether there is a light rail station.
In fact, light rail stations and their immediate surroundings may be safer to be in than on the roads or the open due to the presence of ambient lighting, police, responsible commuters, and surveillance cameras (Jaffe). Increased foot traffic can discourage potential criminals from committing a crime in the first place. According to Todd Litman of the Victoria Transportation Policy Institute “crime risk declines as more responsible (non-criminal) people live, work, and walk in an area, which tends to justify policies that encourage public access such as well-connected streets and paths, houses and shops close to sidewalks, and walking and cycling encouragement” (Litman 8).
Additionally, access to light rails and other forms of public transport will help to decrease crime by tackling the underlying causes of criminal behavior: unemployment and poverty. For low-income households, personal vehicles may not be economically viable due to the costs of gas and maintenance, and due to the fact that these vehicles may be unreliable as a result of deferred maintenance. However, public transportation manages to address these hurdles by “[increasing] labor participation, particularly by lower-income workers” (CTS 2010; Sanchez, Shen, and Peng 2004) and “for every 10 percent increase in the compact development index, there is a 4.1 percent increase in probability that a child born to a family in the bottom quintile of the national income distribution reaches the top quintile… by age 30” (Ewing and Hamidi 2014). The compact development index measures “the urban compactness of populations, facilities and amenities centered on public transport systems at both community and city levels”. Promoting a higher-quality rail system connecting Baltimoreans to the rest of the metropolitan region would enable residents to access more opportunities for employment. Communities should begin to embrace public transportation nodes as catalysts to economic growth, not a public nuisance.
Urban metros with extensive public transit infrastructure are economically wealthier and have lower rates of crime. When Atlanta expanded the MARTA system with additional stations, many residents in the suburbs protested due to the perception that with public transit, there will be more crime. These fears have turned out to be a myth rather than reality. According to Eric Jaffe of CityLab, a study done by Theodore Poister, professor of public administration at Georgia State University “analyzed crime around two relatively new MARTA stations, Kensington and Indian Creek… The baseline crime rates were notably higher in Kensington, an area with greater commercial development and general activity just inside the I-285 loop that circles Atlanta, than in Indian Creek, a more residential place just outside it” and that “any effect on crime rates… was ‘marginal, rather than dramatic’” (Jaffe). Likewise, with Arundel Mills Mall in Anne Arundel County, crime rates seem to be primarily affected by the types of services available within a locality. In the aggregate data, Arundel Mills Mall was included in the non-transit average and had a very high number of thefts. The more commercialized an area, the more likely it is to attract criminals than transit stations attracting or exporting criminals.
Another major U.S. city to have a controversial light rail system is Los Angeles. When the Green Line was being expanded, many residents feared that criminals would deteriorate their neighborhoods by using the light rail to commit crime and escape easily since “[t]he Green Line corridor passes through communities that are quite different”. These fears stem from the fact that the route of the Line passes through “stations in high-crime inner city areas (e.g. Vermont, Harbor, Avalon, Wilmington, and Long Beach Blvd. stations)” (Liggett, Loukaitou-Sideris, and Iseki 4).
To analyze the concerns about violent and property crime in the LA Green Line corridor, researchers conducted an analysis of crime data. The research methodology used in their analysis was similar to the one used in the Transportation Alliance’s, charting points on a GIS map, drawing a half-mile radius around the stations, analyzing the crime rates in those areas, and comparing those rates before the stations opened. The study concluded that “[t]ransit has certainly not brought more crime to the affluent suburban areas, which have continued to enjoy relatively higher levels of safety and prosperity than the county average”. The study proposed an alternate theory on increasing crime rates in L.A. as being influenced by land usage, rather than transit itself. Analyzing hotspots around the county showed that the presence of large commercial centers and liquor stores were more likely to increase the incidence of such activities. The researchers ultimately failed to find evidence that criminals “used the Green Line to access potential targets, miles away” (Liggett, Loukaitou-Sideris, and Iseki 11).
While data for Baltimore City and Baltimore County was unavailable at the time of this post, I will apply the same methodology once the respective police departments are able to send the reports to the Transportation Alliance. So far, empirical studies and analysis of the Atlanta and Los Angeles metros have been unable to find a definitive relationship between transit and increased incidence of crime. Rather, these trends seem to be primarily influenced by the services present in an area, which corroborates our findings in Anne Arundel County. I plan to update the results once I have analyzed Baltimore County and Baltimore City.
Below are images detailing the stations and the controls that are being compared from Anne Arundel County. Note when drawing a circle with a half-mile radius, there will be overlapping radii with some of the addresses, particularly between the Nursery Road and North Linthicum stations. As a result, there may be some double counts in the crime data. Moving forward we hope to solve this issue by grouping stations together if they are too close, instead of analyzing crimes for each individual address and having potential double-counts. In the meantime, since the Anne Arundel County data is already available and analyzed, we wanted to point out this possible double counting. Below are the stations (blue) and controls (red):
Works Cited:
1: Ewing, Reid, et al. “Relationship between Urban Sprawl and Physical Activity, Obesity, and Morbidity — Update and Refinement.” Health & Place, vol. 26, Mar. 2014, pp. 118–26, doi:10.1016/j.healthplace.2013.12.008.
2: Liggett, Robin, Anastasia Loukaitou-Sideris, and Hiroyuki Iseki. “Journeys to Crime: Assessing the Effects of a Light Rail Line on Crime in the Neighborhoods.” Oct 2002. 1 Feb 2009
3: Litman, Todd. 2014. A New Transit Safety Narrative. Journal of Public Transportation, 17 (4): 114–135.
4: Jaffe, Eric. “The Myth That Mass Transit Attracts Crime Is Alive in Atlanta.” CityLab, 17 Dec. 2014
5: Tomas Sanchez, Q. Shen and Z. Peng. “Transit Mobility, Jobs Access and Low-income Labour Participation in US Metropolitan Areas,” Urban Studies, Vol. 41, №7, pp. 1313–1331.
