«A Journal of Policy Development and Research HoPe VI Volume 12, Number 1 • 2010 U.S. Department of Housing and Urban Development Office of Policy ...»
Townes at the Terraces Townes at the Terraces is located on the west side of Baltimore (hereafter referred to as West Baltimore) in the East Poppleton neighborhood. I relied on the Census Bureau to define study neighborhood boundaries (that is, census tracts). As shown in exhibit 2, the microneighborhood (the area immediately surrounding the redevelopment, as explained in more detail below) is bordered to the north by the U.S. Route 40 ramp, to the east by Martin Luther King Jr. Boulevard
(a major road), to the south by the Hollins Market neighborhood (tract 1803), and to the west by the West Poppleton neighborhood (tract 1802). It replaced the public housing project called Lexington Terrace, which consisted of 667 units (housing 2,100 people) in five 11-story highrise and 22 lowrise buildings (Raffel et al., 2003). The buildings were razed in 1996 and, by 1999, the new lowrise HOPE VI redevelopment was open for occupancy. The new HOPE VI townhouses include 391 new units: 250 public housing units, 41 market-rate rental units, and 100 subsidized homeownership units. In addition, a 4-story building with 88 units for seniors housing and an office building with retail space were constructed as part of the redevelopment (HABC, 2006). An elementary school was demolished along with the highrises in 1996, and plans were to reopen it in 2004 (Raffel et al., 2003); however, the site remains a vacant lot to this day. Poe Homes, a lowrise public housing project to the west of Townes at the Terraces, was renovated before the HOPE VI redevelopment but was not demolished (JHU MPP, 2003).
The results of investment in the Poppleton area in the past decade and a half in general have been disappointing. Aside from the HOPE VI redevelopment, two major revitalization initiatives have been undertaken in Poppleton: an EZ and a University of Maryland (UMD) biotechnology park.
Poppleton was designated as an EZ in 1994.5 This federal program funded tax incentives and grants to stimulate economic development in the area. Although this zone encompasses the HOPE VI neighborhood, much of the EZ’s economic activity is focused on the UMD neighborhood across The City of Baltimore received an EZ grant of $100 million in 1994. This grant was allocated among six geographic areas in the city (managed by “village centers” that were created under the EZ program) that totaled 6.8 square miles (City of Baltimore, 2007).
from Martin Luther King Jr. Boulevard (JHU MPP, 2003). Press accounts indicate that Baltimore’s EZ initiative has not been successful in stimulating business growth or developing job opportunities and placement (Anft, 1999, 2000).
In 2002 UMD announced plans for developing its biotechnology park across from Martin Luther King Jr. Boulevard on West Baltimore Street. The project will include 10 buildings for lab and office space and 2 parking garages on 10 acres of land. Construction began in 2004, and, at the time of this writing, one building is complete and occupied and the second building is almost complete.
Although the HOPE VI redevelopment was not an explicit reason for UMD to cross Martin Luther King Jr. Boulevard—cited by some as a symbolic divider between the more developed UMD neighborhood to the east and the more distressed neighborhood to the west—removing the highrise public housing buildings may have contributed to making way for this investment. City and community leaders have expressed high hopes for this revitalization effort by UMD (Beamon, 2004); however, it is still too early in the project’s life to assess any neighborhood spillover effects.
Heritage Crossing Heritage Crossing, also located in West Baltimore, is north of U.S. Route 40 and just southwest of Pennsylvania Avenue, a street with some commercial activity (exhibit 3). The neighborhood on the other side of Pennsylvania Avenue is Upton, known for its rich African-American heritage from decades ago but now, like the Poppleton neighborhood south of U.S. Route 40, an area with high crime and many vacant homes. Heritage Crossing is a sprawling 32-acre development consisting of 75 public housing units and 185 subsidized homeownership units. It replaced the four 14-story Murphy Homes public housing buildings, containing 781 units.
Expectations for Heritage Crossing’s spillover potential are mixed. The development is very large and appears isolated from the surrounding community. Although it has been able to attract low- to moderate-income homebuyers for its row homes, its inclusion of only public housing units and homeownership units has led to tension between the new homeowners and the returning public housing residents. Also, investor speculation was a major factor in this neighborhood (Seipp,
2007) and throughout Baltimore in the early 2000s (Dewar, 2003; Dolan, 2005). Although news of redevelopment piqued the interest of outside investors and attracted private investment into the neighborhood, it has also led to irresponsible investment decisions that may have hampered neighborhood improvements prompted by the HOPE VI redevelopment.
Broadway Overlook Broadway Overlook sits at the northern edge of the Washington Hill neighborhood in East Baltimore. The original 22-story highrise and 14 lowrise Broadway Homes public housing buildings were located diagonally across the street from the new Broadway Overlook HOPE VI development.
The Johns Hopkins Medical Institutions (JHMI), consisting of the Johns Hopkins University Hospital and its medical campuses, are just north of the new development and have a major presence in the neighborhood. During the HOPE VI planning process, JHMI made an agreement with HABC and the Broadway Homes residents to swap the land where Broadway Homes formerly was located with the land where Broadway Overlook is now situated. These locations are shown in exhibit 4.
in the Broadway Overlook HOPE VI project.6 The new development consists of 166 new residential units: 84 public housing units, 5 subsidized homeownership units, 48 market-rate rental units, and 29 market-rate homeownership units.
Interviews with city experts and local press accounts reveal a positive and hopeful impression of the Broadway Overlook redevelopment (Brophy, 2006; Seipp, 2007; Shea, 2006). Because Broadway Overlook is the fourth HOPE VI development in Baltimore, the developer and the other stakeholders involved in its planning had the advantage of being able to learn from the experiences of the previous three HOPE VI projects. The Broadway Overlook architects made a concerted effort to integrate the building design into the diverse architectural landscape of Washington Hill (Gunts, 2003). In addition, the private developer of Broadway Overlook managed the project’s community and social services, whereas HABC had controlled these in the other two study sites. Broadway Overlook’s developer and tenant organization formed a partnership and created effective programs for employment and family support that dramatically increased the tenants’ employment levels and median income (Shea, 2006). The developer also involved the tenant organization extensively in the planning process. Finally, Broadway Overlook has a wide range of housing types, including subsidized and market-rate rentals and subsidized and market-rate homeowner properties, which may help avoid tensions between homeowners and public housing residents. With nearly 30 percent of the units being market-rate rentals, there may be an added incentive to the project’s management to maintain the property and keep it attractive to market-rate tenants. Public housing units are scattered throughout the development and are indistinguishable from market-rate units (Gunts, 2003).
The Broadway Overlook redevelopment also differs from the Townes at the Terraces and Heritage Crossing redevelopments because it is located in a neighborhood where revitalization efforts were already under way. In Washington Hill, homeownership and renovation initiatives were funded by the Maryland Department of Housing and Community Development and implemented through local organizations. Home prices were starting to increase, and JHMI was bringing investment into the community before the HOPE VI initiative. Therefore, it is difficult to be sure that observed improvements in neighborhood conditions are attributable to the removal of an element of blight in an otherwise up-and-coming neighborhood—thus allowing for the acceleration of neighborhood revitalization and improvement—or to positive spillovers from the HOPE VI redevelopment itself.
Data Sources Data for this study come from the Baltimore Policy Project7 and include the address, price, and date of all property sales in Baltimore City from 1990 through the end of 2006. These data were supplemented with data from Maryland Property View 2005, which includes structural characteristics of Baltimore City properties such as year of construction, lot size, structure size, number of JHMI invested $3 million in the HOPE VI redevelopment and other resources targeted at the surrounding community (JHU MPP, 2003).
Professor Sandra Newman of the Johns Hopkins University graduate program in public policy developed and maintains this database.
Neighborhood Definition A central aspect of this study’s methodology is examining the extent to which price levels in the area immediately surrounding the HOPE VI sites deviate from price levels in the same neighborhood but farther away from the sites. Created for the study, microneighborhoods for the three HOPE VI sites served as impact areas around the projects. The first step in creating the microneighborhoods was to use geographically coded sales data to identify which sales fall Missing values in these data are imputed using a multiple imputation method. See Appendix A for a detailed description of how missing values are handled.
Arms-length sales best reflect market values of properties. They exclude transactions between related parties (for example, spouses, relatives, and affiliated companies).
within a certain distance from each HOPE VI site. Like the Holin et al. (2003) study, this study used a single ring to define the microneighborhood for each project. For Townes at the Terraces and Broadway Overlook, the study used a 1,500-foot ring. For Heritage Crossing, which covers a substantially larger area, the study used a 2,000-foot ring.10 Macroneighborhoods are full census tracts that contain property sales within this ring; therefore, the macroneighborhoods include sales within the microneighborhoods and sales outside them but within the same census tract.
For the two HOPE VI sites in West Baltimore, Heritage Crossing and Townes at the Terraces, the
study further defined these neighborhoods to adjust for two additional neighborhood boundaries:
(1) a highway ramp that cuts through the area south of Heritage Crossing and north of Townes at the Terraces and (2) a major road, Martin Luther King Jr. Boulevard, that runs along the eastern boundary of both neighborhoods.11 All property sales south of the highway were excluded from the Heritage Crossing neighborhood, and all property sales north of the highway were excluded from the Townes at the Terraces neighborhood. In addition, few of the sales within 1,500 feet of Townes at the Terraces were located on the opposite side of Martin Luther King Jr. Boulevard.
Therefore, these sales, along with the other sales in that census tract, were also excluded from the Townes at the Terraces neighborhood.
Previous studies used rings of 500 to 2,000 feet, sometimes further defined by 500-foot gradients (for example, Galster, Tatian, and Smith, 1999; Santiago, Galster, and Tatian, 2001). Studies of place-based interventions generally used rings of 1,000 or 2,000 feet (for example, Ellen and Voicu, 2006; Holin et al., 2003). I chose the ring size based on the size of the development.
The Baltimore City Mayor’s Office of Neighborhoods and the Baltimore Neighborhood Indicator Alliance also use these boundaries to define neighborhoods. The Poppleton/Terraces neighborhood lies south of U.S. Route 40, while the Upton/ Heritage neighborhood lies north of it, and the Seton Hill and UMD neighborhoods lie east of Martin Luther King Jr.
Timing This methodology requires identifying periods before and after intervention. It would be ideal to examine prices and trends during the time up to the grant award announcement, between the announcement and demolition, between demolition and project completion, and after completion.
Although these milestones could potentially affect property values in the surrounding neighborhood, insufficient observations are available to reliably capture trends during each of these phases.
The study used the project completion date as the intervention date, expecting that the greatest effect would occur after the project was completed. Because it is possible that the physical removal of the highrise public housing projects encouraged investment in surrounding properties and thus increased property values, the study tested the sensitivity of the main results, using the demolition date as an alternate intervention date. The study did not test the HOPE VI grant announcement dates, because they occurred very early in the study period and too few sales were completed before the announcement to produce reliable estimates.12 Results from JHU MPP (2003) show evidence of very limited spillover effects for the announcement dates of the three redevelopments included in this study.