Rebound Neighborhoods in Weak Market Regions:
The Case of St. Louis
Todd Swanstrom, University of Missouri-St Louis
Hank Webber, Washington University in St. Louis
March 17, 2014
Prepared for Presentation at the Urban Affairs Association Annual Conference, San Antonio, Texas,
March 22, 2014
1
Neighborhoods in American cities are changing all the time. A study of 35 metropolitan areas
from 1950 to 2000 found dramatic change in the economic status of neighborhoods, with the change in
relative economic status of census tracts ranked within each metropolitan area averaging roughly 12 to
13 percent up or down per decade (Rosenthal, 2007). Researchers have extensively studied the causes
and consequences of neighborhood decline. Research on revitalizing or rebounding neighborhoods is
less extensive but growing. A recent study of over 50,000 census tracts between 1970 and 2005-2009
classified between 13.6 percent and 20.6 percent in each decade as “ascending” (Owens, 2012). 1
Most research on revitalizing neighborhoods views them as instances of “gentrification,” the
movement of young, often single, professionals into low-income, usually minority, neighborhoods near
urban employment centers. Scholars have long debated the costs and benefits of gentrification. (For
overviews of the gentrification debate, see Lees, Slater, and Wyly, 2008; Brown-Saracino, 2010). The
dominant view in the literature has traditionally been that in gentrifying neighborhoods the new, higher
income residents end up displacing the low-income and minority residents and taking over the local
culture and consumption patterns. The implication is that gentrification harms previous low-income
residents of the area, who, pushed out of the neighborhood by rising rents, are forced to sever their
social ties and often end up paying more for replacement housing.
Recently, scholars have presented a more complex and nuanced picture of gentrification.
Despite its negative connotations, gentrification has the potential to improve conditions for long-term
residents. By introducing higher income residents into high-poverty areas, gentrification can improve
conditions for low-income residents who remain (Freeman, 2006). A study of New York City found that
low-income residents were actually less likely to move out of gentrifying neighborhoods than other
neighborhoods (Freeman, 2002). Gentrification can improve job prospects for low-income residents,
improve public services by upgrading the tax base, generate more retail opportunities, and even
improve the financial health of existing residents (Vigdor, 2002; Hartley, 2014).
Researchers have mostly studied gentrification in strong market or rapidly growing regions, such
as Boston, Seattle, and San Francisco. Here we focus on revitalizing neighborhoods in St. Louis, a weak
market, or slow growing, region. In this article we ask the question of whether neighborhood
improvement in St. Louis has resulted in substantial and damaging displacement for existing residents,
or whether it has benefited lower income and minority residents. Do all rebound neighborhoods fit one
of these two extremes or are there, in fact, many different types of rebound neighborhoods?
We are also interested in the causes of neighborhood revitalization; in particular, the causes
that potentially can be controlled by local actors. What are the necessary and sufficient conditions of
neighborhood revitalization and which of those factors can local governments, nonprofits, and residents
control? Can local actors influence the trajectory of revitalizing neighborhoods in ways that protect the
well-being of low-income and minority residents who live in the area?
Theories of Neighborhood Revival: Economic, Sociological, and Political/Institutional
Theories of neighborhood change have mostly grown out of research on neighborhood decline.
Recently, however, there has been a growing scholarly literature on the phenomenon of gentrification,
or the revival of older neighborhoods. In this section we review theories of neighborhood change. In
many instances theories developed to explain neighborhood decline can be turned on their heads to
explain neighborhood revival.
2
Theories of neighborhood change come in three basic forms, drawing on the insights of
economics, sociology, and political science: 1) economic theories based on rational actors in free
markets; 2) sociological theories that stress the importance of status concerns, racial attitudes, and
networks; and 3) political explanations that focus on the role of public policies and powerful institutions
in shaping neighborhood outcomes. After reviewing these three theoretical approaches for explaining
neighborhood revival, we offer hypotheses for testing in St. Louis.
Economic Theories of Neighborhood Revival
Early economic theories of neighborhood change originated with the Chicago School of human
ecology in the 1920s. According to Burgess’s concentric zone theory, high-income households move
into new housing on the urban periphery, leaving older housing behind for lower income families (Park,
Burgess, and McKenzie, 1925). The outward movement of high-income households is driven by the
different demand curves for housing by high- and low-income consumers. According to Alonso’s bid
rent theory, high-income households are willing to trade off longer commutes to jobs in the central
business district in order to inhabit larger homes on large lots in the suburban fringe (Alonso, 1960).
Homer Hoyt modified Burgess’s concentric zone theory, developing what he called the sector theory of
neighborhood change: cities tend to develop outward from the center along transportation corridors;
once established in one sector, high-rent neighborhoods “tend to move out in that sector to the
periphery of the city” (Hoyt, 1939, p. 119). Economic theories portray neighborhood change as an
orderly process in which high-income households move out to new housing in the suburbs, setting in
motion a housing filtering chain by which older housing gradually passes down to middle and lower
income households as higher income households migrate outward.
Early economic models were good at explaining why older urban neighborhoods gradually
moved down the economic ladder, but they were not good at explaining why some older neighborhoods
reverse course and move up socioeconomically. In their 1959 study of the New York metropolitan area,
Edgar Hoover and Raymond Vernon noted that some high-income professionals were violating the
classic filtering process which would predict their preference for large houses in the suburbs. As early as
the 1950s, Hoover and Vernon discovered that some higher income professionals were moving back into
older urban neighborhoods. They explained this early gentrification as driven by a desire to live closer
to white-collar employment centers in Manhattan (Hoover & Vernon, 1959). The implication is that as
suburban commutes lengthen, eventually some higher income professionals will prefer to live in smaller
housing units on more expensive land in order to enjoy shorter commutes to work.
Economic supply and demand factors driving rebound neighborhoods can be summed up as
follows: 2
Demand Factors
 Expanded professional employment in downtowns drives up demand for housing in nearby
neighborhoods.
 Growth in the number of small, childless households and retirees increases demand for smaller
urban housing, such as brownstones, condominiums, and luxury apartments.
 Consumer tastes shift in favor of amenity-rich, pedestrian friendly neighborhoods.
Supply Factors
3

Longer commutes from suburban housing make living in close-in urban neighborhoods more
attractive.
 A supply of undervalued older urban housing with attractive architectural features attracts inmovers.3 (This suggests that the in-movement of higher income households will follow the same
wedges or transportation corridors that guided outward migration in earlier periods.)
 Urban land close to employment centers left vacant and abandoned at the end of the filtering
process offers attractive opportunities for investors (Rosenthal, 2007).4
The strength of the supply and demand factors driving neighborhood revitalization will vary
from one metropolitan area to another. One study estimated that a third of variation in neighborhood
conditions can be attributed to metropolitan-level factors (Weisbourd, Bodini, and He, 2009).
Neighborhood revitalization will be more difficult in metropolitan areas in which housing market
demand is being siphoned off from older parts of the region by new suburban development. As David
Rusk puts it, the “inside game” of neighborhood revitalization needs to be supplemented with an
“outside game” that puts limits on suburban sprawl (Rusk, 1999). Metropolitan areas with relatively
slow population growth, often associated with low immigrant populations and few barriers to outward
development, such as deserts, mountains, and oceans, will make neighborhood revitalization more
difficult by increasing the supply of alternatives to older urban housing. The decentralization of
employment out of the central business district will also weaken the demand for older urban housing.
Economic theories can provide coherent explanations as to why some neighborhoods, after
suffering decades of downward economic decline, enjoy renewed consumer and investor confidence.
Based on the assumption that markets tend toward equilibrium as supply and demand are constantly
adjusting, economic theories portray neighborhood decline and renewal as a gradual and orderly
process. Economic theories clearly capture some of the underlying dynamics of neighborhood change,
but anyone who has studied neighborhood change knows that it is far from a gradual and orderly
process. Neighborhoods can go up and down very quickly and land values can fall off a cliff at the
boundary between two neighborhoods. To account for the sudden and often chaotic processes of
neighborhood revival, we must turn to social and political explanations.
Sociological Theories of Neighborhood Revival
According to economic theory, households and investors make independent decisions based on
what is in their rational interest. In contrast, sociological theory postulates that actors are motivated by
cultural values and norms passed on from one generation to the next and expressed through social
structures and roles. According to sociologists, people do not just act as independent rational actors;
they are acutely interested in status hierarchies and social networks and they make decisions on where
to live and invest accordingly.
Sociological theories of status hierarchies based on race, ethnicity, and economic class have
been used to understand neighborhood change.5 In a nutshell, according to sociological theory, people
at the top of social hierarchies prefer to live among others like themselves.6 Thus, whites prefer to live
among other whites. These processes of self-segregation have contagion effects and tipping points that
can set in motion neighborhood trends that are difficult to reverse (Grodzins, 1957; Schelling, 1969;
1971). Thomas Schelling demonstrated that under a wide range of preferences for different racial
mixes, neighborhoods will tip toward completely white or completely black – even though no one
4
intends this result. The fact that the distribution of races in American neighborhoods is not a normal
curve but is more U-shaped is prima facie evidence for racial tipping points. Robert Sampson reports
that from 1960 to 2000 not a single neighborhood in Chicago transitioned from predominantly black to
predominantly white, supporting the notion of a universal tipping point at 50 percent black (Sampson,
2012, p. 107). A recent review of the evidence on integrated neighborhoods in American cities paints a
more complicated picture (Ellen, Horn & O’Regan, 2012). The number of all-white neighborhoods has
declined significantly; many more neighborhoods are diverse in the sense of having a mix of whites and
other races. On the other hand, primarily black communities continue to struggle to attract white
residents.
Status-seeking through residential location is also pursued by the rich who prefer to live among
other affluent households. Beginning with William Julius Wilson’s original 1987 work on concentrated
poverty, researchers have documented the rise of economic segregation in American neighborhoods
and, in particular, the disturbing rise of concentrated poverty.7 The flip side of concentrated poverty is
concentrated affluence. As wealthy households move into a neighborhood, it can become more
attractive for other high-income households, setting in motion a process of rising housing and land
values costs that pushes out lower income residents.
The implication of economic theory is that neighborhood revival will take place in a gradual and
smooth manner. With supply and demand for housing kept in rough equilibrium the socioeconomic
slope between neighborhoods will be gentle. Under concentric zone theory, for example, the urban
land gradient, or underlying land prices, should decline slowly and evenly as you move outward from
employment centers. Neighborhood change should also be slow and steady as the market constantly
adjusts to the changing preferences of consumers and investors.
Sociological theory, with its reinforcing processes of social sorting, can help explain why the
urban land gradient, far from being a smooth gradually sloping plain, is often characterized by jagged
peaks and valleys. Land prices can drop precipitously from one neighborhood to another. Also,
neighborhood change can happen very rapidly, as when panic selling sets in. The concept of tipping
points based on reinforcing cycles can explain the sudden and discontinuous nature of much
neighborhood change. Ascending neighborhoods may reach a tipping point, swiftly displacing the
previous low-income and minority residents. Instead of supply and demand being in equilibrium, under
certain conditions a neighborhood can become “hot” as a high status location and, as a result, prices
increase rapidly. Home purchases may be motivated as much by a desire for speculative gains as by the
inherent characteristics of the neighborhood. Depictions of gentrifying neighborhoods as resulting in
wholesale displacement of low-income and minority residents are rooted in sociological processes of
cumulative causation.
Sociological theory also suggests that the urban land gradient can be characterized by peaks and
valleys. Hot, or rapidly rising, neighborhoods can exist next to very cold, or depressed, neighborhoods.
If homebuyers make decisions partly on the bases of racial and ethnic status hierarchies, which in
America are still generally correlated with economic class, then housing demand can become
concentrated in predominately white neighborhoods and higher income households will shun housing in
a nearby predominantly minority area even if objectively the two areas are very similar.
The dominant prediction from the theory of tipping points is that as neighborhoods rebound
and attract higher income and “higher status” racial and ethnic households, they will tend to tip into
5
becoming all or majority white and upper income communities. Economically, racially, and ethnically
diverse stable rebound neighborhoods will be rare. The degree to which this tipping point process
actually occurs in rebound neighborhoods is an empirical question. Our study of rebound
neighborhoods in St. Louis provides data for evaluating tipping point theory.
The ability of neighborhoods to rebound from urban decline also depends on the actions of
governments and large institutions that are able to shape local housing markets. Uneven neighborhood
development can either be accentuated or ameliorated by political/institutional interventions. We now
turn to this third source of neighborhood change.
Political/Institutional Theories of Neighborhood Revival
According to the political/institutional approach, the principal drivers of neighborhood change
are not individuals or social norms and networks but powerful institutions, especially governments.
Public policies have always played an important role in neighborhood change. In earlier periods
governments played a direct role in neighborhoods; in more recent decades powerful private and
nonprofit institutions have partnered with governments to shape neighborhood trajectories. The ability
of neighborhoods to draw strength from political and institutional relationships varies significantly
across metropolitan areas.
The classic case of public policy driving neighborhood decline is the federal government’s home
loan guarantee programs. In the 1930s, Homer Hoyt was hired to advise the new federal loan guarantee
programs. Hoyt’s job was to provide advice on how to reduce the risk of guaranteeing mortgages. As
we discussed in the previous section, Hoyt’s research showed that there was a racial and ethnic status
hierarchy and people shunned neighborhoods that included low-status groups such as blacks or
southern Italians. Partly on the basis of Hoyt’s research, the federal loan guarantee programs
discriminated against black and racially diverse neighborhoods. The unwillingness of the federal
government to guarantee mortgages in diverse urban neighborhoods exacerbated the tipping point
phenomenon discussed earlier and contributed to the decline of many inner-city neighborhoods. The
loan guarantee programs also discriminated against mixed-use neighborhoods on the grounds that
home purchasers preferred all-residential areas. The federal urban renewal and highway building
programs also forcibly displaced many urban residents, overwhelmingly impacting low-income and
minority households. The roiling of the social structure and invasion of neighborhoods by forcibly
displaced households contributed to neighborhood decline.
The decline of many urban neighborhoods was also driven by public policies that subsidized
suburban development. The tendency of large water, sewer, and electric utilities to subsidize new
development on the backs of existing rate payers made it less expensive for residents of older urban
neighborhoods to flee to outlying suburbs. Autonomous suburban governments passed zoning laws
that excluded rental housing and required expensive single-family homes on large lots. This meant that
the inner-city poor could not follow jobs out to the suburbs. Rising rates of poverty and fiscal stress in
cities contributed to neighborhood decline. Redlining by banks also contributed to neighborhood
decline. According to growth machine theory, landowners join forces with investors and governments
to form a “growth coalition” that promotes commercial development, often at the expense of
residential neighborhoods (Logan & Molotch, 1987). In the 1970s and 1980s political scientists
documented a “corporate-center strategy” by local governments that focused on downtown
development at the expense of neighborhoods (Fainstein, et al, 1986).
6
Although many actions of institutions and governments contributed to neighborhood decline,
another set of policies was designed to support the revitalization of older urban neighborhoods.
Government policies in support of older neighborhoods have shifted significantly in the post-industrial
period. Direct government funding of housing has dropped significantly. HUD production of housing
fell from 248,000 units in 1977 to only 18,000 in 1996 (Erickson, 2009: xiii). The inventory of public
housing is shrinking. The Community Development Block Grant (CDBG), the main successor to the War
on Poverty programs, has shrunk to only about $4 billion for the entire nation. Instead of directly
constructing housing, the federal government and states have assembled a series of tools that can be
used for neighborhood revitalization. The primary federal housing production tool is the Low-income
Housing Tax Credit (LIHTC), which subsidized construction of more than 2 million units between 1987
and 2006 (Erickson, 2009, 148).
In order to use the new policy tools for neighborhood revitalization, local actors must assemble
networks of actors (Keyes, et al, 1996). Community development corporations (CDCs) are key actors in
the new policy paradigm. The first CDCs grew up in the 1960s with the help of Senator Robert Kennedy.
Today there are almost 5,000 across the nation. CDCs engage local residents, devise strategic
neighborhood interventions, and pursue funding to implement them. But CDCs cannot do this work
alone. They need intermediaries to develop their capacity and syndicate the tax credits to wealthy
investors. The capacity of community development systems varies tremendously across metropolitan
areas. National intermediaries and foundations, such as LISC, NeighborWorks, Living Cities, and
Enterprise Community Partners, work in various cities across the country to build up the capacity of the
community development system. Cities without these national connections are disadvantaged in the
competition for funds for neighborhood revitalization. The participation of local community
foundations is also a crucial ingredient in building local capacity to revitalize low-income neighborhoods
(Lowe, 2006).
So-called “anchor institutions” are key actors in the new decentralized community development
system. Anchor institutions are defined as institutions that are tied to specific locations “by reason of
mission, invested capital, or relationships to customers or employees….” (Webber and Karlstrom, p. 1).
Universities and large medical complexes (“eds and meds”) in particular are becoming increasingly
important players in metropolitan economies. Between 1991 and 2011, while manufacturing jobs fell
31 percent, jobs in education and health services increased 87 percent. Eds and meds have little
mobility and many have come to realize that their well-being, especially their ability to attract talented
professional employees, depends on the success of the neighborhoods around them. Many anchor
institutions are investing heavily in neighborhood revitalization. Neighborhoods with strong anchors are
in a much better position to “rebound” from neighborhood decline.
In sum, the political/institutional approach predicts that the ability of a neighborhood to
rebound will depend on its political power and institutional relationships, as well as broader public
policies. It is not enough that a neighborhood be close to large employers; it must also partner with
these anchors to bring resources from outside the neighborhood to the task of neighborhood
revitalization. The ability of a neighborhood to form local nonprofits that steward redevelopment is
also important, as is its ability to access and utilize public policies, from governmental grants to zoning
laws, that support neighborhood redevelopment.
Data and Methodology
7
Our research is focused on two questions: 1) Why do some neighborhoods rebound in the wake
of urban decline while others continue to decline or stagnate? 2) Do rebound neighborhoods fit a
common pattern or do they vary in significant ways both in the path to revitalization and the impact on
previous residents of the neighborhood and surrounding areas?
Our unit of analysis is the “neighborhood.” We define a neighborhood as an identifiable section
of a city where social networks are stronger within rather than across neighborhood boundaries and
where residents identify with the area (Schwirian, 1983, p. 84). Neighborhoods are often defined by a
common history and by political boundaries, such as wards. In the absence of data collected by
neighborhood geography, we use census tract data to trace neighborhood change in the St. Louis
metropolitan area over the forty years from 1970 to 2010. In order to ensure that we are tracking
uniform geographies across time, we utilize the US2010 Longitudinal Tract Data Base (LTDB), which
normalizes data for each census into 2010 tract boundaries.8
Although the U.S. Census makes an effort to follow neighborhood boundaries and natural
barriers when delineating census tracts, census tracts are not the same as neighborhoods. In most
cases, we are able to combine census tracts in ways that approximate commonly understood
neighborhood boundaries in the City of St. Louis, and municipal boundaries sometimes function like
neighborhoods in suburban parts of the region.9 Given our interest in how older neighborhoods
rebound from decline, our data base consists of all 218 census tracts in the “urbanized area” of St. Louis
as defined by the U.S. Census Bureau in 1950 (Figure 1).10 The parts of the St. Louis metropolitan area
that were primarily developed after 1950 are not included in our analysis.
Figure 1.
In order to track neighborhood decline and rebound we devised a three-part Neighborhood
Vitality Index: 1) Economic (per capita income in constant 2012 dollars); 2) Social (percentage of
population not in poverty); 3) Physical (percentage of occupied housing units). The Neighborhood
Vitality Index is a tract-level simple additive index of standardized scores (Z-scores) for these three
variables. For each variable, a standardized score (Z) is computed by subtracting the variable’s mean
value ( ) from the variable’s observed value ( ) and dividing by the standard deviation (s). Expressed
symbolically:
8
( - )/s
The resulting standardized scores are then summed, so that for every tract:
NVI= Zi+Zp+Zo
Where NVI is the Neighborhood Vitality Index, Zi is the Z-score of per capita income, Zp is the Z-score of
non-poverty rates, and Zo is the Z-score of occupancy rates. This calculation provides a measure of how
the census tract did relative to the mean score for all 218 census tracts for that year.
We define a “Rebound Tract” as any census tract that moved up at least 10 percentile points in
the rankings from 1990-2000 or 2000-2010.11 Because we were not interested in neighborhoods that
were always strong and got stronger, we eliminated tracts that were never in the bottom half of the
distribution at some point between 1970 and 2000.12
Running Up the Down Escalator: Older Neighborhoods in Regional Context
The ability of older neighborhoods to revitalize depends on the broader regional context. A
third of the variation in neighborhood outcomes has been attributed to metropolitan-level changes
(Weissbord, Bodini, and He, 2009; see also Jun, 2013). St. Louis is an older industrial region or what is
sometimes called a “weak market” region. Regional job growth, wage growth, and population growth
have been modest. Overall, the St. Louis metropolitan economy has performed less well than
comparable regions. For example, on median household earnings and earnings per job, in 2009 St. Louis
ranked 23rd out of 35 peer metropolitan areas (East-West Gateway, 2011). A weak regional housing
market has been defined as a market where the ratio of median household income to median housing
price is less than 3:1. For the St. Louis metro area the ratio in 2012 was 2.95:1 (American Community
Survey, 2013). 13 But the more important facts in St. Louis concern not the overall strength of the
regional job and housing markets but the distribution of jobs and housing across the urban and
suburban landscape.
The economic performance of the older urban core has been undermined by the flight of jobs
and investment to the suburbs. St. Louis ranks 6th out of the largest 100 metropolitan areas in the
percentage of jobs located more than three miles from the central business district (CBD) (Kneebone,
2013). The demand for housing around downtown is weaker in St. Louis than in other comparable
metropolitan areas. Jobs in the St. Louis region are located in a number of clusters stretching from
downtown west in what is called the central corridor.14
Population has also thinned out significantly in the St. Louis metropolitan area. Figure 3 shows
the growth of the urbanized land area from 1950 to 2010, which has grown much faster than the
population.15 Like other weak market metros around the country, in St. Louis the construction of new
housing units has consistently exceeded the growth in new households (Figure 4).16 The inevitable
effect of constructing an excess supply of housing has been vacant housing at the end of the filtering
chain in older parts of the region. As Figure 5 shows, despite the demolition of tens of thousands of
housing units, the housing occupancy rate has fallen consistently in the older parts of the region, falling
faster than the metropolitan area as a whole.
9
Figure 2. Job Clusters in St. Louis Region
Figure 3. Growth in Urbanized Land Area, 1950-2010
10
Figure 4. Ratio of New Housing Units to New Households
Ratio of New Housing Units to New Households
for St. Louis, MO-IL MSA
1.60
1.427
1.40
1.20
1.274
1.080
0.991
1.00
0.80
0.60
0.40
0.20
0.00
1970-1980
1980-1990
1990-2000
2000-2010
Source(s): U.S. Decennial Census 1970-2000; American Community Survey 2006-2010
The decentralization of jobs and housing, along with overproduction of housing, has put a
tremendous strain on the older neighborhoods in the region which are basically competing over a
continuously shrinking pie. The outlying suburban and exurban areas have enjoyed population growth
at the expense of the older areas (Figure 6). The older parts of the region are, in effect, running up a
down escalator. Neighborhoods have to compete against other neighborhoods for a shrinking pie to
avoid being drained of vitality by the centrifugal forces in the region.
The strength of the centrifugal forces draining older neighborhoods, however, has clearly
declined over the past forty years. In the 1970s the older parts of the region lost an astounding 20.4
percent of their population. Each decade since then the rate of population loss in the older parts of the
region has slowed down (Figure 6), suggesting the emergence of counter-trends favoring older
neighborhoods.
11
Figure 5. Occupied Housing Units by Decade
Percentage of Occupied Housing Units in St. Louis, MO-IL MSA and Study
Area by Decade
96.00%
94.00%
93.90%
93.23%
93.60%
92.60%
92.00%
91.93%
91.90%
90.2%
90.00%
89.12%
88.71%
88.00%
MSA
Study Area
86.00%
84.97%
84.00%
82.00%
80.00%
1970
1980
1990
2000
2010
Source(s): U.S. Decennial Census 1970-2000; American Community Survey 2006-2010
Figure 6. Change in Population of MSA and Study Area
Change in Population of MSA and Study Area by Decade
140.00%
120.00%
118.17%
Percentage of 1970 Popula on
110.18%
100.00%
100.00%
80.00%
99.72%
103.43%
79.59%
71.38%
MSA
64.88%
60.83%
60.00%
Study Area
40.00%
20.00%
0.00%
1970
1980
1990
2000
2010
Source(s): U.S. Decennial Census 1970-2000; American Community Survey 2006-2010
Trends Favoring Older Neighborhoods in the St. Louis Region
After suffering calamitous declines in the 1970s and 1980s, many central cities are enjoying
revivals. A main cause of this revival is a shift in the dominant living patterns of rich and poor. Alan
Ehrenhalt calls this “the great inversion.” “The truth is,” he writes, “we are now living at a moment in
which the massive outward migration of the affluent that characterized the second half of the twentieth
century is coming to an end” (Ehrenhalt, 2012, p. 7). As suburban commutes have lengthened, more
and more affluent households are looking for pedestrian-friendly neighborhoods closer to work. The
revival of central city neighborhoods is also being driven by demographic shifts, particularly the growth
of singles, retirees, and childless households, for whom large single-family suburban homes are less
attractive. The housing preferences of these smaller households are more attuned to apartments,
condos, and row houses – especially those with easy pedestrian access to urban amenities, such as
restaurants, bars, and coffeehouses. Figures 7a and 7b show the growth in the number of single-person
households, married couples without children, young adults (18-34) and elderly (65+) in the St. Louis
region. Clearly, demographic groups that are more inclined to prefer urban living are growing.
12
Figure 7a. Single Households and Married Households without Children, 1970-2010
Figure 7b. Young Adults (18-34) and Elderly (65+), 1970-2010
Central city neighborhoods are well-positioned to compete for the changing housing
preferences of affluent households and the growth of smaller households. Childless households are less
concerned about under-performing central city schools and more affluent households often send their
children to private schools. Older neighborhoods have greater access to public transit, especially light
rail. Older neighborhoods tend to have a greater diversity of housing types than newer suburbs and
feature smaller units that are more attractive to single and childless households. Most importantly, the
older parts of the St. Louis region were largely built out before zoning codes separated uses and,
therefore, provide a mix of land uses that creates the potential for a pedestrian friendly environment
(Figure 10). Not only do older neighborhoods offer a diversity of housing types and therefore economic
diversity, but they are more likely than the exurbs to have the racial and ethnic diversity that many
Millennials find appealing.17
13
Figure 10. Walkability of St. Louis Neighborhoods
Source: http://www.walkscore.com/MO/St.Louis
Findings on Rebound Neighborhoods
Figure 8 is a map showing the relative ranking on our index of neighborhood vitality of each
tract in our study area between 1970 and 2010. The height of each bar represents its ranking in a
particular decade (the higher the bar, the higher the ranking among the 218 census tracts). Focusing on
rebound neighborhoods, our goal was to identify tracts that rose up from a low ranking or had a ushaped trajectory, declining and then coming back up in the rankings in more recent years. Figure 9
highlights the identified rebound neighborhoods.
The first thing to notice about our rebound tracts is that they are heavily concentrated in or near
the central corridor where the three main job clusters in the region are located (Figure 2).18 This
supports one of the key hypotheses derived from economic theory that gentrifying or ascending
neighborhoods will be located near clusters of professional employment. The central corridor is where
higher income households in St. Louis originally resided (directly west of downtown escaping the
congestion and pollution of industry that had spread north and south along the Mississippi River).
Outward migration of the affluent in the twentieth century also proceeded directly west along this
corridor as predicted by Hoyt’s sector theory of neighborhood change (Hoyt, 1939, ch. IV). Historically,
the central corridor is where the most expensive historic housing was built. The impressive brick homes
left behind provide a supply of architecturally valuable housing for reinvestment. As the affluent
migrated west along the central corridor they also left behind an impressive array of urban amenities
and institutions, most notably Forest Park, Washington University, Saint Louis University, and Barnes
Jewish Hospital. Built before zoning codes separated uses, the central corridor has many pedestrianfriendly neighborhoods (Figure 10) and MetroLink, the light rail system, runs down the center of the
central corridor, increasing its connectivity.19
14
Figure 8. Census Tract Rankings by Decade, 1970-2010
Figure 9. Rebound Census Tracts
15
Economic theory predicts that neighborhood revitalization will be driven by increased demand
for urban living among the growing number of single and childless households as well as elderly “empty
nesters.” Our data analysis (Figure 11) shows that rebound neighborhoods did have significantly higher
percentages of young people and single households than non-rebound census tracts. Rebound tracts
actually had lower percentages of elderly and married couples without children. Neighborhood
revitalization in St. Louis is being driven by young single households. The reputation of the St. Louis
public schools has undoubtedly contributed to the low number of households with children under the
age of 18 in our rebound census tracts.20
Figure 11. Residential Demographics in Rebounding and Non-Rebounding Tracts
Tipping point theory would suggest that rebound neighborhoods which attract more highincome professionals and white residents will reach tipping points at which most low-income and
minority residents will be displaced. This idea of a reverse tipping point, based on the in-migration of
affluent whites, underlies the classic concept of gentrification as leading to widespread displacement of
previous residents. To test this idea, we constructed a racial diversity index that measures the
probability that two individuals randomly selected from a single census tract will be of different races.
Essentially, this attempts to measure the degree to which a place facilitates interaction among
individuals of different races – assuming that racial interaction is random and based on spatial
propinquity. Our measure is similar to an isolation index in measures of metropolitan segregation. Our
racial diversity index was computed using the following formula based on four racial groups:
D = 1 - w2 + b2 + h2 + o2
Where D= racial diversity, w= % white non-Hispanic, b= % black non-Hispanic, h= % Hispanic and
o= % other races. The highest diversity score possible is .75 (25 percent in each racial category); the
lowest possible score is zero (entire population of one race).
16
We did not find evidence that rebound neighborhoods in St. Louis reach tipping points that drive
them toward economic and racial homogeneity. To our surprise, our rebound tracts had a mean
diversity score of .40 and a median score of .49, compared to mean and median scores for nonrebounding tracts of .30 and .27. Our rebound neighborhoods are more racially and ethnically diverse
than the older neighborhoods that did not rebound.
Our rebound neighborhoods also do not appear to be subject to a tipping point where inmigration of higher income professionals leads to the wholesale displacement of lower income
households from the neighborhood. We computed a score for each census tract based on the
percentage of residents in professional occupations and the percentage of residents in poverty, in both
cases using census data. We used this measure of economic diversity to provide a preliminary test of
the idea that the in-movement of young professionals forces out the poor. Compared to non-rebound
neighborhoods, our rebound neighborhoods were much more likely to have higher than median poverty
rates and higher than median residents with professional occupations (Figure 13). In other words,
many of our rebound neighborhoods have more professionals and poor people, contradicting the idea
that professionals are gentrifying the neighborhoods and pushing out all the poor people.
Figure 12. Diversity Index Scores, Rebounding and Non-Rebounding Tracts
Figure 13. Economic Diversity of Rebounding and Non-Rebounding Tracts
17
Lacking panel data that follows households over time, our analysis does not enable us to
determine if longtime minority and low-income residents have been able to remain in rebound
neighborhoods. The continued racial and economic diversity of rebound neighborhoods in the St. Louis
metropolitan area strongly suggests, however, that there has not been large-scale involuntary
displacement. This finding contradicts the classic image of gentrification. Our findings are consistent
with a study of six corporately sponsored redevelopment areas in St. Louis which concluded that the
areas became more economically and racially diverse after redevelopment. “[T]he improvements they
have seen look very little like gentrification” (Monti & Burghoff, 2012, p. 528).
There are a number of possible reasons why the rebound census tracts in our study remained
relatively diverse.
1) The weak housing market in St. Louis may have meant that housing inflation in rebound
neighborhoods was not sufficient to displace most low-income residents.
2) When rents did increase, the availability of subsidized housing may have enabled lowincome and minority residents to remain.
3) The presence of a diversity of housing types, e.g., modest apartments mixed in with
expensive single-family homes, may have made it easier for low-income persons to remain
in rebound neighborhoods.
4) Levels of racial intolerance among the new residents of rebound areas may have been low
enough to prevent a racial tipping process to have been set in motion.
5) Civic groups may have mediated relations across racial and economic divides, increasing
levels of trust and toleration, and short-circuiting racial tipping points.
6) These results may reflect a fundamental change in the attitudes of some portion of the
population. Young adults in particular are showing greater preference for living in racially
and economically diverse neighborhoods, and the successful rebound neighborhoods may
be capturing a population that actively values diversity as a neighborhood attribute (see
endnote 15).
Sorting out these different hypotheses will require further research.
To sum up, our quantitative analysis of rebound neighborhoods provides broad support for
economic theory which predicts that revitalizing neighborhoods can be explained by changing market
demand for urban living and supply of attractive housing stock near employment clusters. Sociological
tipping point theory, which predicts that revitalizing neighborhoods will be subject to a reverse tipping
process, is generally not confirmed by our analysis. This should not be taken to imply that residential
choice in St. Louis is not influenced by racial and economic status hierarchies. With some exceptions,
relatively few all-black neighborhoods rebounded, even those located right next to rebounding
neighborhoods. St. Louis is still one of the most racially segregated regions in the United States (though
segregation has been slowly declining for forty years).21 And the number of people living in
neighborhoods of concentrated poverty is increasing in the St. Louis metropolitan area.22 What our
research does suggest, however, is that older urban neighborhoods that revive often are racially and
economically diverse.
Although our analysis generally confirms that rebound neighborhoods are driven by economic
factors, especially proximity to clusters of high-wage professional employment, there are exceptions to
this pattern. As our map of rebound neighborhoods (Figure 9) shows, rebound neighborhoods are
18
scattered throughout the older urbanized area, some located far from employment clusters. Moreover,
some neighborhoods located close to employment clusters are not rebounding, especially areas north of
Delmar Boulevard, the historical dividing line between black and white St. Louis. At the same time,
surprisingly, a few neighborhoods in all-black North St. Louis have modestly rebounded.
We believe the variation from classic economic theory can be primarily explained using the third
theoretical tradition, political/institutional theory. Some neighborhoods have been much more
successful than others in forming high-capacity local organizations to access the resources of
government and anchor institutions in the area. In order to explore the variety of institutional and
political resources that have been mobilized to facilitate neighborhood revitalization and to identify
practical applications of economic theory, we conducted case studies of five very different rebound
neighborhoods.
Case Studies
The purpose of the case studies was to understand the causes of neighborhood change and to
identify best practices for local actors seeking to improve neighborhoods. Each case study provides
stories of localized change and the individuals and institutions responsible for that change (full case
studies can be found in ________ by Webber and Swanstrom).
The case study neighborhoods were selected specifically for their differing characteristics. The
five neighborhoods vary by location (central corridor, northern and southern St. Louis, suburb); historic
economic status (from historically wealthy to working-class); and current character (popular with
professionals, attractive to students and young couples, working class).
Neighborhood Performance, 1970-2010
Using the Neighborhood Vitality Index as a proxy for overall neighborhood performance, Figure
14 provides an overview of the relative performance of our five case study neighborhoods. The diversity
of performance seen in 1970 has persisted to 2010, although each neighborhood has seen varying
degrees of success over time.
Figure 14. Performance of Case Study Neighborhoods 1970-2010
250
Index Score
200
Central West End
Botanical Heights
150
Shaw
Maplewood
100
Mark Twain
Study Area Mean
50
Study Area Median
0
1970
1980
1990
2000
2010
19
Central West End
The Central West End is located in the heart of the central corridor of St. Louis, adjacent to
Forest Park (one of America’s great urban parks), the MetroLink light rail system, the Washington
University Medical Center, and a major regional highway. Due to its location and excellent housing stock
the Central West End was historically among the most attractive and affluent neighborhoods in the city
of St. Louis, but the neighborhood was hit hard by suburban flight in the mid-1900s. By 1970, low
occupancy rates and a high poverty rate (somewhat inflated by a large student population) had driven
the Central West End below average in the Neighborhood Vitality Index. Throughout the 1970s the
neighborhood struggled to deal with high-quality but aging historic housing and declining success in the
neighborhood’s commercial district, including the closure of a grand historic hotel that dominates the
neighborhood’s skyline. Despite these challenges, by 1990 the Central West End had improved greatly as
the result of a variety of factors including: substantial investment by large and growing anchor
institutions; strong citizen leadership; careful contextual physical development; utilization of Historic Tax
Credits and Low Income Housing Tax Credits; and the revitalization of Forest Park. The Central West End
today has a high Neighborhood Vitality Index score, reflecting a 68% increase in per capita income
between 1970 and 1990 and a further 12% increase in per capita income from 1990 to 2010, and is seen
as one of the most attractive neighborhoods in St. Louis.
Botanical Heights
Botanical Heights is located near the central corridor in south St. Louis City. The neighborhood,
previously called McRee Town, is a historically blue collar neighborhood made up of modest single- and
multi-family homes. McRee Town was formerly part of the Shaw neighborhood, but became physically
and politically separated from Shaw in 1973 when an interstate highway was built separating the two
neighborhoods. Although McRee Town was historically white, its white population dropped from 99% to
48% between 1970 and 1990. By the 1990s, McRee Town had become notorious as one of the most
dangerous and depressed neighborhoods in St. Louis and had the seventh-lowest Neighborhood Vitality
Index score in our study area. Following a transformational redevelopment project led by the nearby
Missouri Botanical Garden and local leaders, Botanical Heights has been reborn as an attractive
neighborhood popular among students, young couples, and working class families.
Shaw
Shaw is located directly south of Botanical Heights in south St. Louis City, and borders on the
Missouri Botanical Garden (one of the three great botanical gardens of the world) and Tower Grove Park
(an attractive and prominent historic park). Shaw was traditionally a predominantly white, middle to
upper-middle class neighborhood featuring strong housing and some economic diversity. During the
1970s and 1980s, Shaw faced a combination of white flight and aging housing, a problem that was
amplified by redlining practices that reduced access to mortgage and home improvement loans in the
area. Through the 1990s Shaw also suffered from increases in crime that mirrored the decline of nearby
Botanical Heights. These struggles are reflected in a substantial drop in the Neighborhood Vitality Index,
but Shaw successfully rebounded above its 1970 score by 2010, with per capita income increasing by
65% between 1990 and 2010. Improvements in Shaw were primarily due to grassroots community
efforts from a neighborhood improvement association and the local Catholic church. In addition, the
transformation of Botanical Heights reduced some of the crime problems that had been bleeding into
20
Shaw and other surrounding neighborhoods. Because of its strong housing stock and nearby amenities,
Shaw has become once again a desired neighborhood attracting a diverse population.
Mark Twain
The Mark Twain neighborhood is located in northern St. Louis City, surrounded by some of the
most devastated neighborhoods in the region. Mark Twain was historically a blue collar community that
housed workers for the many nearby manufacturing jobs. Prior to the 1960s, the population was
primarily white and middle class, but with the disappearance of manufacturing industries in north St.
Louis and the onset of white flight, Mark Twain’s population quickly became entirely African American
with high levels of poverty. While Mark Twain has not rebounded to the extent of the other case study
neighborhoods, its success has been in stabilizing while all of its neighbors continued to decline. While
poverty rates remain high and the educational level of residents is very low, Mark Twain has retained
high occupancy rates and saw a 7% increase in per capita income between 2000 and 2010, reflected in a
rebounding Neighborhood Vitality Index score. The stabilization success in Mark Twain has been driven
by two small local anchors and the work of resident leaders, aided by relatively strong and manageable
housing stock.
Maplewood
Maplewood is an inner-ring suburb of St. Louis, located directly southwest of the city but with
easy access to central corridor amenities, including the MetroLink light rail. Maplewood is a classic
streetcar suburb, featuring a variety of single-family and multi-family homes suitable for a mix of lowerand middle-income residents. The neighborhood as a whole was historically above average in the
Neighborhood Vitality Index, and has remained so for the past 40 years, although a set of challenges
caused one of its census tracts to drop below average in 1990. Maplewood’s challenges stemmed from
declining quality of housing, a failing commercial district and school district, and mismanagement in
government. Successful responses to these challenges included concerted efforts to attract small
businesses to the commercial district, a city-wide dedication to improve the school district, and an
overhaul of the government.
Diversity
Despite the wide range of outcomes found in our case study neighborhoods, each of these
neighborhoods has seen substantial improvement while generally maintaining or increasing levels of
resident diversity. Every neighborhood saw a loss in white population in the 1970s and 1980s as white
flight hit the core of St. Louis. Despite that, by 2010, other than Mark Twain, none of these
neighborhoods was made up of more than 75% of a single race.
The one exception to this rule is Mark Twain; since the 1990s it has remained an entirely African
American community. This is true of most of the neighborhoods in northern St. Louis, and follows the
predictions of traditional sociological theory that neighborhoods tend to tip toward all white or all black.
Of the 51 census tracts in our study area that were under 10% white in either 1970 or 1990, none
reached even 15% white by 2010.
21
Figure 15. Racial Makeup of Case Study Neighborhoods 1970-2010
100
Central West End
% non-white
80
Botanical Heights
60
Shaw
Maplewood
40
Mark Twain
Study Area Mean
20
Study Area Median
0
1970
1990
2010
Success Factors
The case studies allowed us to identify eight factors that increased the likelihood of success in
the rebounding neighborhoods. All of the case study neighborhoods displayed some of these factors,
but the number and strength of factors varied greatly. Success in a neighborhood did not rely on the
existence of all of these factors, but a higher quantity and intensity of the success factors increased the
probability of success.
Figure 16. Success Factors in Case Study Neighborhoods
Central
Botanical
Success Factor
Shaw
Mark Twain Maplewood
West End
Heights
Strong Anchor Institutions
X
Excellent Housing Stock
X
Thoughtful Commercial
Development
Thoughtful Residential
Development
X
X
(X)
X
X
X
X
Resident Civic Engagement
X
X
X
Good Location
X
X
X
Successful Public Policy
X
Strong Public Schools
X
X
X
X
X
X
(X)
X
Anchor Institutions
Political theory supports the idea that local anchor institutions have emerged as key players in
the new decentralized community development system. The need for investment from anchors is
amplified further in a city with a weak tax base and a state that has not historically invested in urban
22
development. Four of the five case study neighborhoods saw significant investment from local anchors.
In the Central West End, several institutions of the Washington University Medical Center (including the
Washington University School of Medicine and Barnes, Jewish, and Children’s Hospitals) chose to not
only remain in the neighborhood in the 1970s while other businesses left, but also actively invested in
improving the safety, livability, physical assets, and attractiveness of the community. These institutions
have grown into the second largest job center in the region. Mark Twain is not located near any major
institutions, but was heavily supported by two small anchors; Bellefontaine Cemetery and New Sunny
Mount Baptist Church. These institutions provided funding for rebuilding of homes, organization of
residents, and social service delivery for residents, particularly children. Their impact in Mark Twain
emphasizes the importance of local anchor institutions, even those that are not as large and wealthy as
hospitals or universities.
Excellent Housing Stock
As a simple matter of economics, neighborhoods with higher quality housing stock have greater
potential to attract residents and investment. The Central West End is the one case study neighborhood
with truly extraordinary housing stock, featuring one of the most impressive historic districts in the
country. There is no doubt that this physical asset contributed to the neighborhood’s ability to entice
wealthy residents back to the city. Shaw’s housing, while not as grand as the Central West End’s, is also
very strong and played an important role in attracting middle- and upper-income families to the
neighborhood.
Thoughtful Commercial and Residential Development
Economic theory also supports the value of thoughtful development of commercial areas
(creating demand for the neighborhood) and residential areas (improving available supply). In addition,
many of these communities have successfully retained a level of economic diversity through
development of a diversity of residential options. Maplewood’s success is due in great part to the
development of its commercial centers, including a downtown full of small unique businesses and a
major shopping center on the edge of town anchored by international chains like Walmart and Lowe’s.
The success of these developments was due to both Maplewood’s purposeful pro-business policies and
a bit of luck in attracting highly successful businesses. The Central West End restored its commercial
district as the city’s center for boutique stores and nightlife, bringing increased vitality to the
neighborhood. Mark Twain was successful in building new housing options that improved the
attractiveness of the neighborhood while remaining manageable in size and cost for working-class
residents.
Resident Civic Engagement
The engagement of local residents is the backbone of classic community organizing. Every case
study neighborhood was supported by an active citizen base, suggesting that this is one factor that is
necessary for success. Rather than relying on government or major institutions to independently
support neighborhood change, community organizing attempts to generate a new source of power
through a critical mass of active residents. In Shaw, improvements were supported by a neighborhood
association and grassroots organizing through the local Catholic church. Botanical Heights’ citizens were
engaged by the Missouri Botanical Garden as part of their community redevelopment planning process.
In addition, a new successful charter school in Botanical Heights was created by a group of local parents
who wanted to stay in the area but had no quality school options for their children.
23
Good Location
Economic theory makes it clear that location is a key factor in attracting residents to a
neighborhood. The definition of a “good” location varies by region and perspective, but for these
purposes the central corridor of St. Louis was viewed as the most advantageous location in the city. This
area contains some of the largest job centers in the region and provides access to most of the major
regional amenities, including Forest Park, the MetroLink light rail system, and the major regional
highways. All of our case study neighborhoods with the exception of Mark Twain are located in or very
close to the central corridor, which significantly improved their ability to rebound.
Successful Public Policy
In order to ma imize a neighborhood’s potential, local actors must be able to successfully
navigate and utilize public policies that can support local development. This ability is often contingent on
a neighborhood’s access to individuals or organizations with the sophistication and knowledge required
to navigate these policies. The Central West End was remarkably adept at utilizing public policies to
support development, especially in physical development. The development done by the Washington
University Medical Center Redevelopment Corporation was aided by Missouri Chapter 353, a state
statute that provides incentives like tax abatement and power of eminent domain to developers of
blighted areas. Residential development was also subsidized by the use of State Historic Tax Credits
(Missouri has one of the country’s largest such programs), and Federal Low Income Historic Tax Credits,
which were particularly helpful in maintaining a level of economic diversity in the neighborhood.
Maplewood successfully used the legislative process to pass multiple bond and tax increases that funded
the transformation of the school district.
Strong Public Schools
There is no shortage of literature regarding the value to neighborhoods of strong public schools.
Without strong school options, a neighborhood significantly reduces its ability to attract families with
children. St. Louis City, which is home to all of the case study neighborhoods except Maplewood, has
had a particularly hard time maintaining school quality, and successful neighborhoods in the city are
succeeding despite the failing public schools, mostly by attracting students and couples without
children. The one case study neighborhood that saw significant success in this area is Maplewood, a
suburb of St. Louis City which has its own independent school district. Maplewood’s rebound was
supported by the transformation of its school district into one of the most successful, diverse, and
innovative districts in the St. Louis region. Botanical Heights has had a taste of success with the
emergence of a new high-achieving charter school that serves only local residents, but the majority of
the neighborhood remains dependent on a set of poor public school options.
Conclusion
The period from 1970 to 2010 was a difficult time for St. Louis, particularly for the older parts of
the region. The region lost important employment centers, a significant amount of the population, and
the core was depleted by white flight and suburbanization. However, the worst part of this period seems
to have ended and the decline of the region is at minimum slowing and may even be reversing. Against
this backdrop, a set of neighborhoods have successfully rebounded, with four of the five case study
neighborhoods achieving success while maintaining high levels of diversity. This success required a
combination of inherent neighborhood advantages (location, thriving local anchors, etc.) and very
careful attention by key local and regional actors. It remains to be tested whether improving
24
neighborhoods while avoiding the negative effects of gentrification is common in other cities, or
whether in stronger markets gentrification is inevitable even if it is unintended.
25
References
American Community Survey 2008 to 2012 (5-Year Estimates), U.S. Census Bureau.
Alonso, W. 1960. A Theory of the Urban Land Market. Papers and Proceedings of the Sixth Annual
Meeting of the Regional Science Association: 149-157.
Bier, T. and C. Post. 2003. Vacating the City: An Analysis of New Homes vs. Household Growth. In A.
Berube, B. Katz, and R. D. Lang, eds., Redefining Urban and Suburban America. Washington, D.C.: The
Brookings Institution Press.
Brown-Saracino, J. 2010. The Neighborhood that Never Changes: Gentrification, Preservation, and the
Search for Authenticity. Chicago, IL: University of Chicago Press.
Bryant, Tim. 2014. Anchors and transit spur growth of St. Louis corridor, St. Louis Post-Dispatch (January
26).
Downs, A. 1981. Neighborhoods and Urban Development. Washington, D.C.: Brookings Institution.
Dreier, P, J. Mollenkopf, and T. Swanstrom, forthcoming. Place Matters: Metropolitics for the TwentyFirst Century, 3rd ed. (Lawrence, KS: University Press of Kansas).
East-West Gateway Council of Governments. 2011. Where We Stand: The Strategic Assessment of the
St. Louis Region, 6th ed. St. Louis, MO: East-West Gateway Council of Governments.
Ehrenhalt, A. 2012. The Great Inversion and the Future of the American City. New York: Alfred A. Knopf.
Ellen, I. G., K. Horn & K. O’Regan. 2012. “Pathways to Integration: E amining Changes in the Prevalence
of Racially Integrated Neighborhoods. Cityscape: A Journal of Policy Development and Research, Vol.
14, No. 3: 33-53.
Erickson, D. J. 2009. The Housing Policy Revolution: Networks and Neighborhoods. Washington, D.C.:
The Urban Institute.
Fainstein, S. , et al. 1986. Restructuring the City: The Political Economy of Urban Development. AddisonWesley Longman.
Freeman, L. 2006. There Goes the ‘Hood: Views of Gentrification from the Ground Up. Philadelphia, PA:
Temple University Press.
Freeman. 2002. Gentrification and Displacement. The Urban Prospect, Vol. 8, No. 1: 1-4.
Fulton, W., R. Pendall, M. Nguyen, and A. Harrison. 2001. Who Sprawls the Most? The Brookings
Institution Survey Series, July.
Glaeser, E. 2011. Triumph of the City: How Our Greatest Invention Makes Us Richer, Smarter, Greener,
Healthier, and Happier. New York: Penguin Press.
26
Grodzins, M. 1957. Metropolitan Segregation. Scientific American 197: 33-41.
Hartley, D. 2013. Gentrification and Financial Health. Federal Reserve Bank of Cleveland November 6.
Hoover, E. M. and R. Vernon. 1959. Anatomy of a Metropolis. Harvard University Press.
Hoyt, H. 1939. The Structure and Growth of Residential Neighborhoods in American Cities. Washington,
D.C.: Federal Housing Administration. , p. 119
Jun,H. 2013. Determinants of Neighborhood Change. Urban Affairs Review 49 (3): 330-352.
Keyes, Vidal, Schwartz, and Bratt, 1996. “Networks and Nonprofits: Opportunities and Challenges in an
Era of Federal Devolution” Housing Policy Debate, Vol. 7, No. 2, 1996
Kneebone, E. 2013. Job Sprawl Stalls: The Great Recession and Metropolitan Employment Location.
Brookings Institution Metropolitan, Metropolitan Opportunity Series, April.
Lees, L. , T. Slater, and E. Wyly. 2008. Gentrification. London: Routledge.
Leinberger, C. B. 2008. The Option of Urbanism: Investing in a New American Dream. Washington, D.C.:
Island Press.
Logan, J. R. & H. L. Molotch. 1987. Urban Fortunes: The Political Economy of Place. Berkeley, CA:
University of California Press.
Logan, J. R. and B. J. Stults. 2011. The Persistence of Segregation in the Metropolis: New Findings from
the 2010 Census. US2010 Project, March 24.
Logan, J. R., Zengwang Xu, and Brian Stults. Interpolating US Decennial Census Tract Data from as Early
as 1970 to 2010: A Longitudinal Tract Database. Professional Geographer, forthcoming.
Lowe, J. S. 2006. Rebuilding Communities the Public Trust Way: Community Foundation Assistance to
CDCs, 1980-2000. Lnaham, MD: Rowman & Littlefield.
Monti, D. J. & D. Burghoff. 2012. Corporately Sponsored Redevelopment Campaigns and the
Demographic Stability of Urban Neighborhoods: St. Louis Revisited. Journal of Urban Affairs 34 (5):
513-531.
Owens, A. 2012. Neighborhoods on the Rise: A Typology of Neighborhoods Experiencing Socioeconomic
Ascent. City & Community 11(4): 345-369.
Park, R. E., E. W. Burgess, and R. D. McKenzie. 1925. The City. Chicago, IL: University of Chicago Press.
Rosenthal, S. S. 2007. Old Homes, Externalities, and Poor Neighborhoods: A Model of Urban Decline and
Renewal. Journal of Urban Economics 63: 816-840.
Rusk, David. 1999. Inside Game/Outside Game: Winning Strategies for Urban America. Brookings
Institution Press.
27
Sampson, R. J. 2012. Great American City: Chicago and the Enduring Neighborhood Effect.
Chicago, IL: University of Chicago Press.
Schelling, T. C. 1969. Models of Segregation. American Economic Review 59 (2): 488-493.
Schelling, 1971. Dynamic Models of Segregation. Journal of Mathematical Sociology 1: 143-186.
Schwirian, K. p. 1983. Models of Neighborhood Change. Annual Review of Sociology 9: 83-102.
Smith, N. 1986. Gentrification, the Frontier, and the Restructuring of Urban Space. In N. Smith and P.
Williams, eds., Gentrification of the City. Boston: Allen and Unwin.
Smith, N. 1979. Toward a Theory of Gentrification: A Back to the City Movement of Capital, not People.
Journal of the American Planning Association 45, October: 538-548.
Vigdor, J. L. 2002. Does Gentrification Harm the Poor? Brookings-Wharton Papers on Urban Affairs.
Washington, D. C.: Brookings Institution Press: 133-182.
Webber, H. S. and M. Karlstrom. 2009. Why Community Investment Is Good for Nonprofit Anchor
Institutions: Understanding Costs, Benefits, and the Range of Strategic Options. Chicago, IL: Chapin Hall
at the University of Chicago.
Weisbourd, R., R. Bodini, and M. He, 2009. Dynamic Neighborhoods: New Tools for Community and
Economic Development. Living Cities, The National Community Development Initiative.
28
1
Owens defines ascending census tracts as those that increased their rank in the metropolitan area on her
indicators of socioeconomic status by 10 percentile points or more.
2
Supply and demand factors underlying neighborhood revitalization are adapted from Downs, 1981, p. 75. For
recent summaries of the economic factors driving urban revitalization, see Ehrenhalt, 2012; Glaeser, 2011; and
Leinberger, 2008.
3
Neil Smith e plains gentrification as the result of investors jumping in to take advantage of a “rent gap” – the
difference between the actual ground rent received for a parcel of land and versus the potential ground rent given
a more profitable use in upscale housing (Smith, 1979; 1986).
4
Rosenthal (p. 822) notes that housing depreciates at about 2 percent per year and therefore after about 50 years
older urban housing is ripe for reinvestment.
5
We focus here on sociological theories based on status. Sociological theories of social capital and trust are also
important in understanding neighborhood change. The level of “collective efficacy” in a neighborhood can e plain
its trajectory independent of economic factors (Sampson, 2012).
6
Of course, affluent households can benefit economically from living among other affluent households, e.g., in
being able to generously fund local public schools while paying relatively low tax rates. However, this kind of
collusion in residential location, while it can be motivated by economic gain, does not fit the market model which
requires that households make decisions independently of each other.
7
For a synthesis of this literature, see Dreier, Mollenkopf, and Swanstrom, forthcoming).
8
For more information on the data go to: http://www.s4.brown.edu/us2010/Researcher/Bridging.htm. See also
Logan, Xu and Stults, forthcoming.
9
Neighborhoods have not been delineated for all of the older parts of the St. Louis region. The Census Bureau has
divided the City of St. Louis into 79 neighborhoods, but data by neighborhood is not available for all the censuses
in our data set. St. Louis County is divided into 91 municipalities; smaller municipalities function like
neighborhoods but larger municipalities have a number of neighborhoods within them. More than a third of St.
Louis County is unincorporated; in some cases the Census Bureau has identified census designated places that
function like neighborhoods.
10
The urbanized area generally consists of contiguous territory that is part of a metropolitan area of at least
50,000 people that has a density of at least 1,000 persons per square mile. For a more complete explanation of
how the Census Bureau defines urbanized area see U. S. Bureau of the Census, Urban and Rural Definitions,
October 1995; available at: http://www.census.gov/population/censusdata/urdef.txt. We only included census
tracts that were wholly within the urbanized area as of 1950; small parts of the urbanized area in 1950, therefore,
are not included in our data set.
11
A tract that moved up in the 1990s was eliminated if it moved down in the 2000s.
12
Note that our rebound tracts are identified by how well they did relative to all 218 older census tracts; rebound
is not based on absolute performance.
13
Median house value for all owner-occupied units=$159,700 (2012 dollars)/ median household income=$54,109
(2012 dollars).
14
The Central Corridor is generally defined as the area between downtown and the River West to I-170 bordered
on the South by I-64 and on the North by Delmar and Washington.
15
St. Louis is one of the most sprawled out metropolitan areas in the nation. Between 1982 and 1997 the
population of the metropolitan area increased by 6.0 percent at the same time that the urbanized land area
increased by 25.1 percent (Fulton, et al, 2001).
16
Using more accurate housing permit data, Bier concludes that in the 1990s St. Louis built 1.7 units of new
housing for every new household (Bier and Post, 2006).
17
A recent survey of 3,000 individuals examined attitudes toward racial and ethnic diversity. The 160-point
composite openness index measured the degree to which people are open to racial and ethnic diversity. The
youngest age group—Americans ages 18 to 29—reported a mean score of 92, compared to the oldest age group—
Americans ages 65 or older—which scored an average of 80 on the index. Center for American Progress and
PolicyLink, in partnership with the Rockefeller Foundation, Building an All-In Nation: A View from the American
Public Survey, Report and analysis written by Ruy Teixeira and John Halpin, Center for American Progress with
29
Matt Barreto and Adrian Pantoja, Latino Decisions, October 2013; available at:
http://www.americanprogress.org/wp-content/uploads/2013/10/AllInNationReport.pdf.
18
The central corridor is not contains the three largest job clusters, but it is also where most major cultural and
recreational amenities are located. See Bryant, 2014.
19
There are 63,721 jobs paying a monthly wage of at least $3333 within a half-mile of transit stations on the light
rail (MetroLink) system. Source: Longitudinal Employer-Household Dynamics, 2009; compiled by the Public Policy
Research Center, University of Missouri-St. Louis.
20
The St. Louis public schools lost state accreditation in 2007 but gained provisional accreditation in 2012.
21
In 2010 St. Louis was ranked as the 11th most segregated metropolitan area out of the 50 areas studied (Logan
and Stults, 2011).
22
For updated analysis of trends in concentrated poverty, see
(http://www.brookings.edu/~/media/Research/Files/Papers/2011/11/03%20poverty%20kneebone%20nadeau%2
0berube/profiles/41180.pdf).