The topic of suburban expansion is timely for the City of San Antonio, Texas. With an estimated population of 1,469,845, San Antonio currently ranks the second largest city in the state and seventh largest in the United States [1
]. The city will add 500,000 jobs and 500,000 units of housing by 2040, testament to a strong and diverse economy built on healthcare, education, military and tourism. All of this in a city that added 430,000 people in the last ten years [2
]. This growth will increase the population of host Bexar County by 65%, with much of the population influx occurring within the San Antonio city limits [3
]. This demographic expansion would likely lead to significant geographic expansion. The city’s current footprint covers 467 square miles, at the relatively low residential density of 3017 people per square mile [3
] (Figure 1
Our research is already registering the impact of this growth, measured as a function of both construction completions and developed land area. A comparison of the 1990s and 2000s, for example, shows a 165% increase in construction completions and a 185% increase in developed land area (Table 1
). This data reveals the largest relative increases during the post-WWII period and the largest total increases ever.
The dramatic expansion of population and land area has prompted city planners to advance a series of recent policy efforts including a comprehensive master plan, transportation plan, and sustainability plan. These documents outline a plan for growth through 2040. The explicit goal of the Comprehensive Plan, passed in 2016, is to determine the direction and form of future physical growth, distribute the projected population, and guide infrastructural investment [3
]. The Comprehensive Plan recognizes that economic activity in San Antonio is distributed widely across 13 regional nodes, nine existing and four emerging, which together account for half of the city’s employment and non-residential development since the year 2000 [3
]. The Comprehensive Plan also recommends further investment in these nodes, with the hope of transforming them into mixed-use activity centers [3
]. The city is also preparing an $
850-million-dollar bond campaign to fund major infrastructure projects. Together, these two policy efforts will go a long way toward determining San Antonio’s ability to sustain existing environmental, financial, and infrastructural systems in the face of unprecedented local growth.
Our research catalogs the history of suburban expansion in San Antonio in order to help policy-makers and residents make more critically informed choices about where and how to grow. We begin the research in the year 1890, which marks the first time that the city began large-scale expansion beyond its original 36-square mile colonial grid. The ensuing 120 years saw San Antonio transform itself multiple times; stretching its historical footprint in a linear fashion along streetcar routes, multiplying growth in a concentric fashion along automotive ring-roads, and eventually assuming the complex polycentric footprint that the city exhibits today.
While the origins of suburban development in the United States date to the early nineteenth century, it was not until the second half of the twentieth century that suburbs fully eclipsed central cities as the primary locale for residential and economic development in U.S. cities. Mieszkowski and Mills [4
] note that between the years 1950 and 1990, the percentage of residents living within central cities, as opposed to larger metropolitan statistical areas, declined from 57% to 37%, while the percentage of jobs declined from 70% to 45%. Wheeler [5
] uncovers similar trends in his study of six metropolitan areas in the U.S., revealing that 88% of their total metropolitan land area was developed after 1940, a time period when suburban morphologies achieved prevalence. Farris [6
] locates a similar disparity in studies of 22 major central cities, finding that during the 1990s they netted only 5.2% of total new metropolitan housing permits.
The ascendance of suburban morphologies during the second half of the twentieth century has generated continuous and widespread disdain from planners, policy-makers, and architects. Early detractors like Peter Blake [7
] critiqued suburbia on largely aesthetic terms, lamenting the negative visual impact of billboards and power lines on previously countryside views. More recently, opponents have focused their ire on the professed social and environmental deficiencies of suburban development [8
]. This critical discourse, impressive in its breadth, depth, and diversity, generally falls under the rubric of New Urbanism or Smart Growth [9
]. These related movements are bound by a series of design and policy prescriptions, many of which seek to recapture the best qualities of the pre-automotive city. Some of the most prevalent strategies include increasing population densities, re-introducing street grids, expanding mixed-use zoning, and investing in mass transit networks.
The last decade, however, has seen attitudes towards suburbia shift as critics have increasingly come to accept the inherent form and logic of suburban landscapes. Robert Bruegman [13
] constructs a powerful historical argument for this position, reminding us that the geographic decentralization that defines suburban locales is not unique to post-war U.S. cities. In fact, even ancient cities like Babylon, Ur, and Rome included transitional or suburban zones, which by definition fall somewhere between urban and rural conditions.
Joel Kotkin [14
] similarly contends that the expanded decentralized structure of U.S. post-war suburbs is fully capable of accommodating growth in an environmentally and socially productive manner. According to Kotkin, the proper role of policy-makers and designers is therefore to facilitate further growth through more productive policy and formal arrangements within the existing suburban structure. Architect Judith De Jong [15
] stakes out a related position, arguing that suburbs offer the most robust site for new formal and spatial investigations. She is less convinced than Kotkin that the garden suburbs of the early twentieth century offer the best way forward. Instead, De Jong suggests that the greatest potential for growth lies in edge cities such as Tyson’s Corner, Virginia; inner-ring suburbs like Evanston, Illinois; or exurban areas like The Woodlands, Texas.
Kotkin and De Jong both advocate for a position that can broadly be described as New Suburbanism. This emerging paradigm accepts the inevitability and even legitimacy of post-war suburban environments, arguing for a series of strategic interventions to improve these landscapes through the introduction of new forms, programs, and spatial arrangements. This position is largely at odds with the more familiar narratives of New Urbanism and Smart Growth, which assert a greater affinity for the footprint of the pre-Modern pre-automotive city. Still, even urbanists who advocate a return to more traditional urban morphologies recognize that the strategic infill and retrofit of the existing suburban fabric offers a critical way to minimize the negative effects of post-war growth and to accommodate new growth [16
This paper argues that New Suburbanism, as a theoretical framework, can help San Antonio confront its impending population boom in three ways. First, the approach would allow San Antonio to absorb 1.1 million people in the next 25 years, even if the city is unable to fill out its existing urban core. This is not a small consideration, as San Antonio’s recent efforts to advance residential and commercial development in the city’s central core have done little to slow centrifugal patterns of suburban development. In fact, our research indicates that the peak wave of construction during the 2000s, a time period tabbed by former Mayor Julián Castro as the ‘Decade of Downtown’, remained a full 15 miles beyond the historic urban core.
Second, the political climate in Texas, characterized by strong support for property rights and fierce resistance to regulations, means that most attempts to guide or limit the location of growth will encounter staunch and well-funded opposition. Most recently, pro-development sentiments led the Planning Commission to remove binding language from the Comprehensive Plan related to impervious cover and night sky restrictions [19
Third, by focusing on the inherent possibilities of both new and infill suburban development, San Antonio policy-makers will have an opportunity to retrofit and infill the massive quantities of peripheral development that have appeared since 1950.
The research focuses on three specific research questions. First, how did the form of suburban growth shift within Bexar County over the study period? Second, where has suburban growth historically occurred within Bexar County during the study period? Third, what type of suburban growth has San Antonio experienced during the study period, specifically with regard to single-family and multifamily housing?
The work utilizes two broad measures of growth, both of which operate at the metropolitan scale. Such expansive methods seem appropriate in Bexar County, a locale that has not benefitted from significant quantitative urban analysis. The team began by mapping the geographic distribution of suburban growth within Bexar County. For this portion of the work, the project relies on kernel density spatial analysis, which establishes how the concentration of development in Bexar County changed over time. This paper presents the data at the scale of the county, classifying it by decades. Kernel density calculates the density of development (point features) by creating a smooth raster surface. Kernel density spatial analysis allows the research team to detect and display emerging development patterns. The technique is especially useful when considering large study areas like Bexar County, which covers over 1200 square miles.
An excellent precedent for this approach comes from Delmelle, Zhou, and Thill [20
], who use the technique to understand the impact of urban infill on residential density in Charlotte, NC between the years 2000–2009. Their work reveals that density in Charlotte is increasing simultaneously within the city’s core and suburban periphery, suggesting that traditional distinctions between a dense urban core and a diffuse suburban periphery are breaking down.
The second portion of this analysis digs deeper into the spatial data, sorting construction permits by year built and distance from center. The research team organizes this data in chart form, highlighting San Antonio’s chronological shift from a centripetal, or inward pattern of growth to a centrifugal, or outward one. This approach builds on a lineage of work that begins with Blumenfeld [21
], who found that cities expanded in three stages, beginning with slow growth, moving towards peak population growth, and finally entering decline. He notes that the peak wave of suburban development travels at a rate of one mile each decade. Subsequent research efforts pick up on the wave metaphor to describe suburban growth, including Hart [22
] and Pond and Yeates [23
]. The latter offer a five-stage development process that begins with agricultural development and ends with complete urbanization.
Gober and Burns [24
] extend this line of research, examining the size and shape of the ‘urban fringe,’ or outer edge of development, in Phoenix between the years 1990 and 1998. They specifically measure the distance between housing completions at the periphery and those at the metropolitan center. This work focuses on where the urban fringe peaks, how fast it peaks, and on its character and surroundings. Their results suggest that the urban fringe does not constitute a highly concentrated narrow region but rather a complex process that impacts various areas of the metropolitan region at different rates and in dissimilar ways.
] similarly examines the changing form and character of the urban fringe, comparing the location and type of new housing construction in Phoenix between 1990–1999 and 2000–2005. The work charts the quantity of construction completions and distance from the center to conclude that the relentless outward momentum of single-family construction has slowed in recent years, being replaced by infill development. The findings further suggest that the urban fringe does not expand evenly in all directions, as barriers to development impede growth in some areas of the city. It finally reveals that multifamily housing, long associated with dense infill at the historic core, is becoming increasingly prevalent at the urban fringe.
Together, this body of research describes an increasingly complex, peripheral, and decentered pattern of suburban development. The resultant landscape, composed of increasingly unfamiliar spatial and programmatic arrangements, will continue to resist the traditional development solutions offered by New Urbanist and Smart Growth advocates. Within this context, New Suburbanism provides a promising theoretical framework for sustainable growth, one grounded in the physical and political realities of contemporary suburbia.
2. Materials and Methods
The research team began by obtaining the 2015 parcel shapefile for San Antonio from the Bexar County Property Appraiser. This shapefile contained information on property ownership, the year the first building was constructed on the lot, the assessed value of building(s) and lot, land use, and the total acreage for the parcel. The team next secured a shapefile with San Antonio building footprints from the Texas Sustainable Energy Research Institute, extracting the building footprints by processing the TNRIS LiDAR dataset. This provided the total coverage of each building and total acreage of developed land.
The next step involved removing all parcels with a missing year-built date from the study before performing a spatial join of the parcel shapefile and building footprints. This process assigned all of the attributes from the Bexar County Property Appraiser dataset to the buildings. The team then constructed shapefiles for both the parcel and buildings for each decade beginning in the 1890s and ending in the 2000s. This allowed the team to observe development patterns and trends for 14 distinct time periods.
The project considered four land uses including residential single-family, residential multifamily, commercial, and industrial. These land uses corresponded to the state reporting categories as defined by the property classification guide for Texas. The team removed buildings and parcels associated with all other land uses (such as utilities and farmland) from the study. This left a sample size of 609,006 buildings and 475,127 parcels.
Upon securing the data, the team engaged in two forms of analysis. First, in an effort to examine how the distribution of suburban development changed over time, we converted the polygon building footprint shapefiles for each decade into point shapefiles in ArcGIS. The next step involved using the kernel density spatial analyst tool on the point building footprint shapefiles to calculate the concentration of buildings within an area. This created a development surface for each time period. The team utilized total building coverage as a weight in the population field of the kernel density function in order to account for the size of each completed building. The team utilized the default parameters for the kernel density spatial analyst tool. The next step involved creating maps from the results and classifying them using the equal interval method. The team then compared the kernel density maps across decades to determine how the concentration of development changed in San Antonio over time.
In an attempt to more fully understand the centrifugal growth in the city, the team calculated the average distance from the city center, building on work from Gober and Burns [24
] and Atkinson-Palombo [25
]. This involved using the ArcGIS near tool to calculate the distance from the center of each building in the data set to the center of downtown’s Main Plaza, widely recognized as the heart of the historic city. The final step involved calculating the total developed acres (i.e., coverage) and sorting the data by use (residential single-family, residential multifamily, commercial, and industrial) and by decade (1890–2010).
The current research confirms several critical trends: first, since the year 2000 San Antonio’s urban development has become increasingly polycentric; second, the rate of centrifugal growth has increased dramatically since 1990; and, third, multifamily housing is playing an increasingly large role in defining the form and program of the metropolitan area. Collectively, these trends attest to San Antonio’s emergence as a polycentric metropolitan area, where the vast majority of growth is occurring beyond the limits of the traditional city.
While this observation does not distinguish San Antonio among Sunbelt cities, it does represent a bold reconceptualization of the metropolitan region. Prevailing public images of the Alamo City focus almost exclusively on downtown, highlighting the beautifully preserved Spanish colonial grid, Alamo, Riverwalk promenade, and Mission district. This image is magnified by the city’s impressive record of historic preservation, which became even more distinguished in 2015 with UNESCO’s inscription of the San Antonio Missions as World Heritage sites. San Antonio’s historic emphasis on the downtown core has driven tourism and urban policy for decades. Still, the data makes clear that halfway through the ‘Decade of Downtown’, relatively little of the city’s development is actually occurring downtown.
The results of this research in no way diminish the physical and cultural importance of San Antonio’s historic downtown and Missions. Rather, they question the relevance of a concentric growth model in light of persistently scattered development patterns that favor the periphery. Modarres [37
] (p. 120) makes the argument that in order ‘to build a sustainable, polycentric or networked city, we need to re-think our notions of urbanism, urban planning, urban management, and development’. Increasingly, the planning community in San Antonio is taking up this challenge while recasting suburban development as something more than an unfortunate aberration.
The recently completed 2016 Comprehensive Plan successfully challenges dated concentric models that would privilege the historic downtown and central business district. The current plan forecasts that no fewer than thirteen distinct employment centers will be required to absorb the continuing economic boom [3
]. This shift is noteworthy to the extent that it imagines the next San Antonio as a rapidly expanding, polycentric, decentralized landscape that will continue to resist geographic containment. Only by fully accepting this reality can San Antonio begin to generate sustainable growth policies that critically engage land use, transportation, open space conservation, and aquifer preservation.
The Comprehensive Plan’s decision to abandon a concentric model in favor of a polycentric one seems particularly appropriate given the lack of political will to implement growth restrictions, such as the ones that curtailed geographic expansion in Portland, Oregon [38
]. If policy-makers truly intend to guide growth towards a more sustainable outcome, they must assertively engage the suburban periphery. This will necessitate new approaches to land development that incentivize urban infill and density, not just in downtown and inner-ring locations but also in post-war suburbs and even at the emerging edge. This approach will require the development of new suburban prototypes that can increase densities and mix uses without the benefit of traditional urban morphologies. Such an approach may also begin to challenge the conventional wisdom that emphasizes the importance of leveraging existing infrastructures, particularly those with physical proximity to the historical center. While the logic of this position is clear, it does not account for the fact that, historically, most development has occurred at the periphery of the city, a zone that has moved continuously farther from the historical center since 1890 (Table 2
, Figure 3
In a span of several decades, San Antonio has emerged as the fastest growing major city in the U.S., easily outpacing Phoenix and San Diego, its two closest competitors [39
]. This seismic shift is placing unprecedented demand on local ecologies and infrastructures. It is also stretching San Antonio’s civic imagination, requiring policy-makers and residents to confront the unfamiliar terms of contemporary urbanism while relinquishing historic notions of a concentric city that lost currency after the Second World War.
As San Antonio scrambles to accommodate this growth, it must aspire to urban models that expand the physical and conceptual parameters of the Spanish colonial grid. Within this context, the theoretical framework of New Suburbanism becomes relevant, infusing the discourse with new ideas while avoiding the lament that often accompanies discussions of suburban growth. While efforts to preserve downtown and infill the central city remain vital, the next chapter in San Antonio’s story is being written at the city’s periphery and beyond. As San Antonio sets course for the next twenty-five years, it must continue to explore a more dynamic relationship with this emerging geography, conceiving a future sufficiently complex to befit the city’s celebrated past.