Community Development Innovation Review

August 23, 2016

Homeownership and the Stability of Middle Neighborhoods


Homeownership, although far from universal, forms a central part of what might be called a national ethos, in which owning one’s own home is associated with middle-class status and the American dream achieved.1 Yet, a recent article by Ryan Cooper bore the grandiose title “It’s Time to Kill the American Dream of Homeownership.”2 Whether homeownership is or is not a good investment for a middleclass family is not the issue here, although there is a compelling case that it still is.3 The question this chapter will attempt to answer is a different one; namely, what role does homeownership play in the vitality of middle neighborhoods in legacy cities?

This question is particularly timely for a number of reasons. First, as Cooper noted, many support the proposition that homeownership is overrated or irrelevant, or, in the recent words of a respected colleague, “it’s time to get over homeownership.” Second, the years since the bursting of the housing bubble in 2006 and 2007 have shown not only a widely reported decline in homeownership rates nationally, but a significantly greater decline in homeownership rates—and in the absolute number of homeowners—in legacy cities.

If a relatively high level of homeownership is indeed an important factor in fostering neighborhood stability, a different phenomenon—a growing number of single-family homes purchased by absentee investors—should be a source of considerable concern to those who care about the future of middle neighborhoods. My case for this proposition is circumstantial; homeownership is interwoven with many other factors affecting neighborhoods, and, as I will discuss, the pathways by which it affects neighborhood vitality are complex and multifaceted.4 At the same time, I would argue that the case is strong, and that homeownership should be at the forefront of policies and strategies to stabilize or revive urban middle neighborhoods.

At the same time, it is important to stress that arguing for the value of homeownership does not imply that rental housing is unnecessary or that renters are in some fashion second-class citizens and cannot contribute to their neighborhoods. Rental housing is a vital part of any community, particularly those with large numbers of lower-income families for whom homeownership may not be a realistic or desirable alternative. While maintaining a high homeownership rate may be a desirable public policy, policies that focus on homeowners and fail to address both the importance of a sound rental housing stock and engaging renters fully in their communities are as unbalanced as strategies that ignore homeownership entirely.

This chapter is in four sections. The first provides a brief historical introduction to homeownership in middle neighborhoods, while the second discusses the research evidence for the neighborhood effects of homeownership and explores some of the pathways by which those effects are experienced. The third describes the erosion of homeownership in legacy cities and their neighborhoods, including a case study of Trenton, New Jersey, where I have been able to use a unique neighborhood-level data set showing the trends in owner-occupant and investor home purchases from 2006 through 2013. The final sections explore the features of a model that links different homeownership effects to neighborhood change and suggest some policy implications for middle neighborhoods.

The Historical Background

The middle neighborhoods of legacy cities were developed beginning in the late nineteenth century through the early 1960s. They were historically, and remain today with few exceptions, neighborhoods of single-family homes.5 In Camden, Baltimore, and many coastal cities, these homes were row houses, while in Toledo, Detroit, and most inland cities; they were detached houses on small, usually narrow, lots. Homeownership rates in legacy cities from 1920 on were often comparable to or higher than the national homeownership rate (Table 1). By 1930, one-half or more of the single-family houses in most of these cities were owner-occupied.

Table 1
Homeownership Rates in Select Legacy Cities, 1900–1960

Cities with homeownership rates above the national average are shaded.
Source: 1900, 1920, 1930, and 1960 Census of Housing, U.S. Bureau of the Census.

Homeownership rates in the cities in Table 1 grew at a far more rapid pace than the national average from 1900 to the Great Depression and World War II in the 1930s and 1940s, and homeownership was common in urban areas well before the reforms of the New Deal. Between 1900 and 1930, the number of homeowners in Baltimore more than tripled, to 97,000, while the number of renters grew much more modestly, from 70,000 to 92,000. The number of Trenton homeowners also more than tripled, to nearly 15,000, while the number of renters increased by fewer than 2,000. In both cities, the character of the housing stock, mainly single-family row houses, did not change materially. In all likelihood, what was happening was that many rented houses became owner-occupied and that the majority of the new houses built were sold to homebuyers rather than absentee landlords.

Homeownership growth in many cities did not end with the Depression. A number of cities, including most notably Toledo and Detroit in Table 1, saw dramatic increases in homeownership following World War II. Between 1930 and 1960, the number of homeowners in Detroit doubled, to 299,000, while the number of renters barely grew, from 211,000 to 215,000. Clearly, and contrary to widespread belief, the increase in homeownership during the immediate postwar period was not a purely suburban phenomenon.

Although data do not exist to enable one to zoom in on particular neighborhoods in these cities, it is reasonable to assume that middle neighborhoods, being inhabited largely by middle-income families and occupying the middle of the local housing market, had homeownership rates similar to or higher than those shown in Table 1, and that well before World War II, homeownership was already a central element in the character of the typical urban middle neighborhood. As I suggest, both here and in the previous essay in this volume, the recent drastic drop in homeownership in many of these neighborhoods has been a significant factor in their decline.

Neighborhood Effects of Homeownership

With homeownership looming so large in the American ethos, it is not surprising that an extensive body of research exists on its effects, whether in terms of wealth-building, behavior and family outcomes, or neighborhood conditions and dynamics. In this section, I summarize the research findings in five separate areas: residential stability, property values, property condition, social/behavioral factors, and social capital and collective efficacy.

All of this research shares the problem of how to isolate homeownership from other social and economic factors. Although the research, particularly more recent work, typically tries to control for socioeconomic differences between owners and renters, such as income or race, it is more difficult to pin down the extent to which homeownership is affected by self-selection; in other words, whether people who choose to become homeowners have different attitudes or values than people of similar social and economic status who choose not to become homeowners. This may in turn affect their behavior and their effect on their surroundings.7

Although this does not affect the relationship between homeownership and whatever neighborhood feature one is trying to measure, such as stability or civic engagement, it means that one can never be completely certain that one is measuring the effect of homeownership or the effect of some other social factor that is, in turn, linked to homeownership. For that reason, the nature of the pathways through which homeownership exerts its influence, which I address later, becomes particularly important.

Residential Stability

Residential stability in legacy city middle neighborhoods is declining as homeownership declines. Residential stability or turnover appears to be an important element in neighborhood health, with high turnover or “churning” seen as a factor leading to decline8. Homeownership is statistically associated with greater length of tenure; the 2013 American Community Survey finds that the median length of residence for homeowners in their current home is 11 years. This compares with fewer than three years for tenants. The tenure gap is even greater in legacy cities, as shown in Table 2.

Table 2
Average Tenure for Owners and Renters in Select Legacy Cities

Medians for renters calculated by author from grouped American Community Survey data.
Source: 1 year 2012 American Community Survey

Analysts have raised the question of how to separate the impact of homeownership as such from the impact of long-term tenure stability (National Association of Realtors 2006). Some research has found that the effect of homeownership on child outcomes drops significantly when controlling for mobility9 . Thus, in theory, one might be able to achieve outcomes similar to those associated with homeownership by stabilizing the tenure of renters or by fostering intermediate forms of tenure, such as rental with tenure rights or share appreciation, as exist in some European countries. In practice, though, this may not be a realistic option. First, evidence is strong that homeownership improves residential stability independent of other socioeconomic factors10. This may be a function of the greater transaction costs for homeowners associated with moving or it may reflect some of the value or attitudinal changes associated with homeownership, as noted earlier. Second, the magnitude of the tenure gap between owners and renters is so great that it is hard, if not impossible, to conceive of a plausible strategy that would eliminate it. Although some advocates have suggested that a landlord-tenant regime that incorporates security of tenure and rent control would have such an effect, the experience in New Jersey, where security of tenure is enshrined in state law and rent control is legal and widely used, does not support that proposition.11 Increasing tenants’ tenure through legal and economic strategies is a desirable policy objective. It would almost certainly yield significant benefits for tenants and may also yield some potential community benefit. However, it is unlikely in the extreme to be able to substitute for homeownership as a means of fostering neighborhood stability.

It is not enough to encourage families to become homeowners. It is equally or more important to ensure that they become stable, long-term homeowners, and that they do not involuntarily lose their homes through foreclosure, tax delinquency, or other controllable factors12. There is abundant evidence that involuntary loss of homes is severely destructive to both the homeowners and their neighborhoods, potentially exceeding whatever benefits were gained by becoming homeowners in the first place.13

Property Values

The value or sales prices of homes in a neighborhood is arguably the single most direct measure of the economic vitality of a neighborhood. Rising property values are a direct indicator of positive economic change in a neighborhood, and declining values equally directly measure negative change. Because homeowners tend to have higher incomes than renters, it stands to reason that property values would be higher in areas with high homeownership levels. There is considerable evidence, however, that, independently of income, homeownership and property values bear a strong relationship to each other.

A number of studies have found that newly constructed, subsidized housing for owner-occupancy increases the value of nearby homes.14 Although these effects may have as much to do with the replacement of vacant lots or derelict buildings, research has found significant price increases with increases in homeownership rates, even after systematically controlling for both neighborhood and individual characteristics.15 Chengri Ding and Gerrit-Jan Knaap have looked at the converse, finding that the loss of homeowners from Cleveland neighborhoods reduced property values in those areas.16 William Rohe and Leslie Stewart have found that the relationship works in reverse as well; healthy property value appreciation triggers greater homeownership.17 This last point offers insight into an important aspect of the pathways that drive neighborhood effects, the process by which households decide where to buy homes.

Property Maintenance and Condition

The condition and maintenance of properties are important elements in a neighborhood’s stability and health. Although research finds a strong relation between homeownership and property maintenance and condition, it also finds that the relationship is contingent, in the sense that homeowners’ maintenance decisions are strongly influenced by other neighborhood features. Both George Galster18 and Yannis Ioannides19 found that the level of social interaction and social cohesion in a neighborhood significantly influences property upkeep. Put differently, a homeowner’s maintenance and investment decisions are influenced by neighborhood expectations and by what he or she sees neighbors doing. Their findings suggest a possible link between homeownership, property upkeep, and collective efficacy. This would be a fruitful area for further research.

Who owns the home is also important. My research in Las Vegas found a significant difference in property conditions between owner-occupied and absentee-owned properties within the same block or neighborhood20. Figure 3 illustrates the difference in property conditions in Flint, Michigan, for owner-occupied and absentee-owned properties, as well as the effect of higher homeownership rates on the condition of rental properties21 The census tracts shown along the X (horizontal) axis in Figure 3 are organized in order of homeownership rate from low to high. The Y-axis shows the average condition score for properties, using a 4-point scale in which properties in good to excellent condition were scored 1, and dilapidated properties scored 4.

Figure 1
Tenure and Property Condition by Census Tract in Flint Michigan

Source: Analysis by author of data from 2010 Census (homeownership rates), 2012 Flint Parcel Survey (average housing condition) and Genesee County Property Records (homeowner/absentee owner distribution).

Figure 1 support the research findings that neighborhood peer behavior plays a major role in driving maintenance decisions. The higher the homeownership rate, the better properties are maintained and the better their condition. At every point on the continuum, moreover, owner-occupied properties are better maintained than absentee-owned properties, with the quality gap largest in areas where homeownership rates are lowest.

At the same time, one should not infer that the effects seen in Figure 1 are necessarily caused by higher homeownership rates. Higher homeownership rates are associated with higher incomes and higher property values, and it is likely that these effects are the result of the interplay between these (and perhaps other) factors.

Mortgage Foreclosure and Tax Delinquency

A number of studies have found that absentee owners are more likely than owneroccupants to allow their properties to go into mortgage foreclosure. Richard Todd, who studied Cuyahoga County, Ohio, early in the foreclosure crisis found that nearly three times as many non-occupant owners in Cuyahoga County had a foreclosure notice filed on their mortgage by April 30, 2008, than owner-occupants (28 percent vs. 9 percent)22. Even when controlling for such factors as income, borrower’s race, and neighborhood housing values, the foreclosure rate on mortgages to non-occupants was at least double that of owner-occupied mortgages. Other research found that the disparity between foreclosure rates for owneroccupants and absentee owners was significantly greater in the midwestern states where legacy cities are typically located than in Sunbelt states such as Nevada and Florida.23

Little or no published research exists on the relationship between homeownership and tax delinquency, although logic would suggest that the same disparities apply. My work in Trenton, New Jersey, supports that proposition. I was able to use parcel-level data to compare tax delinquency and redemption rates for owner-occupants and absentee owners of single-family homes (Table 3).

Table 3
Percentage of Absentee Owner Properties with Tax Liens on File in 2014 in Trenton, New Jersey

Source: City of Trenton tax collector. Analysis by author.

Table 3 suggests that although the likelihood of early tax delinquency is only moderately greater for absentee owners (+15 percent), the likelihood of long-term delinquency— reflected in the failure to redeem 2012 and 2013 tax liens as of late 2014—is significantly greater (+65-75 percent) for absentee owners than for owner-occupants.

Social and Behavioral Conditions

Many studies find a strong connection between homeownership and different family social or behavioral conditions, and these conditions can affect neighborhood stability in important ways. Changes in child and youth outcomes may affect crime through lower drop-out rates, in turn leading to lower juvenile delinquency; or through lower teen pregnancy rates leading in turn to lower poverty rates in the next generation. These relationships reflect the well-established link between teen pregnancy, single female parenthood, and poverty. Richard Green and Michelle White found a strong relationship between homeownership and greater educational attainment, lower dropout rates, and fewer teen pregnancies.24 Other researchers have found that the children of homeowners are more likely to achieve higher levels of education and subsequent earnings, controlling for other relevant social and economic factors affecting educational outcomes and earnings25. It is likely that a strong feedback chain exists between such behavioral changes at the family level and neighborhood conditions.

Research also has found that homeownership is associated with better physical and psychological health26, overall life satisfaction27, and owners’ greater sense of control over their environments28. The extent, however, to which these factors affect neighborhood conditions remains uncertain.

It should be stressed that these positive effects are the product of successful homeownership, reinforcing the point made earlier that public policy should not aim simply to create homeowners but to foster sustainable homeownership. Homeowners who are delinquent on their mortgages or mired in foreclosure proceedings suffer from increased stress, depression, and mental illness29. The possibility should not be dismissed that these psychological effects contribute to the well-documented powerful negative effects of foreclosure on neighborhood vitality.

Social Capital and Collective Efficacy

Social capital can be seen as a combination of civic engagement and trust or the extent to which people feel mutual obligations to one another (Putnam 1993). Kenneth Temkin and William Rohe studied change in Pittsburgh neighborhoods between 1980 and 1990 and find that “neighborhoods with relatively large amounts of social capital are less likely to decline when other factors remain constant.”30 A related concept linking social dynamics to neighborhood change is collective efficacy, or the “social cohesion combined with shared expectations for social control.”31 This concept echoes a much earlier formulation by Jane Jacobs, who wrote “a successful neighborhood is a place that keeps sufficiently abreast of its problems so it is not destroyed by them.”32

Notably, however, “social control,” Sampson, Raudenbush, and Earls write, “should not be equated with formal regulation or forced conformity by institutions such as the police and courts. Rather, social control refers generally to the capacity of a group to regulate its members according to desired principles—to realize collective, as opposed to forced, goals.”33 They found that collective efficacy is “a robust predictor of lower rates of violence,” after controlling for neighborhood characteristics.34 Later research has found that the absence of collective efficacy to be a strong predictor of homicide rates35.

Homeownership is positively associated with social capital. Homeowners are much more likely to participate in activities that increase neighborhood social capital, such as volunteering or participating in block group meetings.36 Manturuk, Lindblad and Quercia found similar patterns when looking specifically at the behavior of low- and moderateincome homeowners37.

Other research has found strong relationships between homeownership, collective efficacy, and neighborhood crime and disorder38. Lower homeownership, or lower collective efficacy, are both associated with higher levels of crime and disorder. This relationship is again subject to the homeowner having a sustainable mortgage. Two European studies also support the link between homeownership and collective efficacy. A Danish study found a strong association between greater homeownership and lower crime in a neighborhood, while controlling for multiple economic and demographic variables39, while a German study found that homeowners were less willing to accept deviant behavior and more ready to intervene when they observe such behavior40.

In conclusion, the relationship between homeownership and neighborhood change is complex and multidimensional, yet it appears clear that increasing stable, sustainable homeownership can significantly further positive neighborhood change through many different pathways, while a decline in homeownership is likely associated with neighborhood decline.

The Erosion of Homeownership in Legacy Cities

Although homeownership rates in legacy cities tended to parallel and even exceed national trends between 1900 and 1960, the trends have sharply diverged since then. In those cities, homeownership is declining and investor purchases are rising. Given the importance of homeownership to neighborhood health, as described above, this is a problematic trend.

All of the cities shown initially in Table 1 saw their homeownership rates drop after 1960, in some cases sharply, as in Flint or Camden, and in others more gradually, as in Toledo or Grand Rapids (Figure 2). Although homeownership rates have declined nationally in recent years, the long-term national trajectory over that period, as shown in Figure 2, was upward.

Figure 2
Homeownership Rates in Select Legacy Cities, 1900–2010

Source: 1900, 1930 and 1960 Census of Housing; 2010 Census of Population

Figure 2 is somewhat misleading, however, given that it implies that homeownership has been declining since 1960 for all of these cities. Instead, many legacy cities saw continued growth or only modest declines in homeownership rates until the collapse of the housing bubble in 2007, at which point the rate plummeted. Table 4 shows the trends for a cluster of large legacy cities.

Table 4
Change in Homeownership Rates, Select Cities, 1960–2007 and 2007–2013

Source: 1960 Census of Housing, 2007 and 2013 1-year American Community Survey (ACS). The cities included had one-year ACS data available for both 1960–2007 and 2007–2013.

Four of the seven cities in Table 4 saw homeownership growth between 1960 and 2007, modest in most cases, but substantial in St. Louis. Since 2007, all seven have seen sharp declines in both homeownership rates and in the number of owner-occupant households (Table 5). As a whole, these seven cities lost 11 percent of their homeowners, or more than 94,000 homeowner households.

An initial inference might be that the changes in legacy cities are no more than a reflection of the erosion of homeownership nationally during this period. This is incorrect, as not only is the rate of decline in these cities more substantial than the national rate of decline, but the numerical decline is far more substantial, as a percentage of the homeowner base, than nationally. The number of homeowners in these cities is declining at a rate of 1 percent to nearly 3 percent per year in the case of Detroit.

Table 5
Change in Number of Homeowners, Select Cities, 2007–2013

Change in Number of Homeowners, Select Cities, 2007–2013

During this same six-year period, the number of renters increased in each of these cities, in some cases substantially. Even in Detroit, where the total population continued to decline precipitously, the number of renters increased by more than 3,000 households.

Several factors drive this erosion of homeownership, but one factor is clearly the increasingly dominant role of investor-buyers in legacy city housing markets. It is hard to measure this trend with precision, although a comparison of total sales volumes with the number of purchase mortgages in the same community during the same period can provide a rough sense of the trajectory of change.41 Table 6 compares sales volumes with purchase mortgage volumes for three cities between 2006 and 2012. Mortgages declined from 42 percent of sales in Cleveland in 2006 to 20 percent by 2012, and in Pittsburgh from 46 percent to 22 percent. In Detroit, where the market collapse was pronounced, mortgages in 2012 represented fewer than 2 percent of total sales.

Table 6
Ratio of Purchase Mortgages to Total Sales, Select Cities, 2006–2012

Source: HMDA, Boxwood Means data from PolicyMap

At the same time, Table 6 makes clear that total sales volumes also dropped significantly, although to a lesser extent, Pittsburgh, which may have the strongest housing market among major legacy cities, being an exception. This drop in sales volume reflects the severe difficulty that would-be homebuyers have in obtaining mortgages in the post-bubble era; a recent Urban Institute report concluded that “tight credit standards prevented 5.2 million mortgages between 2009 and 2014”42. Although investors have filled part of the gap in effective market demand, much remains unfilled, leading to greater property abandonment in weaker neighborhoods. Moreover, as I have discussed in detail elsewhere, depending on the underlying market conditions of the neighborhood, investor behavior may have significant destabilizing effects.43

My recent study in Trenton, New Jersey, offers a more detailed picture of increased investor activity.44 I analyzed individual sales transactions between 2006 and 2013 to identify investor and homebuyer activity citywide and by neighborhood for each year.45 The trend shows a pattern consistent with that shown by the comparison of sales and mortgage data. The number of sales plummeted, with the number of owner-occupant homebuyers declining from more than 1,000 in 2006 to an average of less than 200 for the past three years (Figure 3). The number of investors has remained relatively stable since 2007 but at a level considerably lower than in 2006, the last year of the housing bubble. In 2013, investors represented nearly 80 percent of all sales in Trenton, compared with 50 percent in 2006.

Figure 3
Sales Transactions by Type of Buyer in Trenton, NJ, 2006–2013

Source: New Jersey sales transaction database. Analysis by author

Although in 2006, the percentage of investor buyers was roughly proportional to their share of the city’s housing stock, by 2013, the investor share was far higher, as illustrated in Figure 4 for two of the city’s middle neighborhoods. Both of these neighborhoods still have relatively high homeownership rates (59 percent in Franklin Park and 64 percent in Parkside). Although investors own only 36 percent of the inventory in Parkside, they have accounted for 68 percent of the purchases there since 2006 and 86 percent since 2011. In Franklin Park, investors own 41 percent of the inventory, but they have accounted for 54 percent of the purchases since 2006 and 74 percent since 2011. The rate of erosion in homeownership in these neighborhoods is likely to be significant.46

Figure 4
Investor Share of Inventory and Purchases, 2006–2013, in Two Trenton Neighborhoods

Franklin ParkParkside
Source: New Jersey real property database and sales transaction database. Analysis by author

Modeling the Relationship between Homeownership Erosion and the Middle Neighborhood

What, then, is the relationship between homeownership erosion and the decline of so many middle neighborhoods in legacy cities? In the previous essay in this volume, I presented data showing the extent of that decline, while in this chapter I have tried to make two points: first, there is a compelling link between homeownership and a host of factors associated with stable, healthy neighborhoods; and second, decline in both the share and the number of homeowners in legacy cities and their neighborhoods has accelerated.47

Although the Trenton study finds a very strong relationship between the investor share of purchases (a reasonable proxy for homeownership erosion) and factors such as median house price, violent crime rate, or tax foreclosure, all of which are associated with neighborhood strength and weakness,48 one cannot necessarily conclude that the decline in homeownership causes neighborhood decline. Nonetheless, there appear to be clear associations between loss of homeownership and decline, and the findings on neighborhood effects suggest a number of the pathways for such a relationship. The balance of this section explores these pathways and suggests a possible model of the relationship between homeownership and neighborhood change.

In doing so, it is essential to distinguish between those effects that appear to be properties of homeownership as such, which may be considered primary effects, and those that are the product of those factors, or secondary (or tertiary) effects. For example, even though there appears to be an association between collective efficacy and homeownership, that association may not be inherent to homeownership in itself, but could be seen as a secondary effect driven by primary features of homeownership, namely the higher level of investment as well as the longer duration of tenure associated with homeownership.

Indeed, stripped to its essence and disregarding the potential of homeownership as a means of building wealth, there are arguably only two salient features that intrinsically distinguish homeownership from rental tenure: the significantly longer duration of the tenure and the fact that homeownership represents a significant financial, and psychological, investment in a place. The two are closely interwoven. Although the financial investment may be independent of the duration of tenure, the psychological investment, to the extent it exists, is likely to be linked to duration of tenure. Duration of tenure, however, may also be linked to financial investment, if only because of the resulting greater “stickiness” of homeownership49 and the higher transaction costs associated with selling a home than renting50.

Figure 5 is a conceptual model of the relationship between homeownership and neighborhood change. The extent to which the specific pathways in the model are supported by the body of research discussed earlier varies widely. The relationships between collective efficacy and crime incidence, or between crime and property values, for example, are strongly supported. The relationship, on the other hand, between length of tenure and collective efficacy is my hypothesis, drawn by inference from the research, rather than a relationship that has been explicitly established by research. Relationships that are more strongly established are shown with bold lines. Although the relationship between homeownership and foreclosure incidence is reasonably well established, the relationship between the financial investment in homeownership and foreclosure is inferred from the prior relationship, rather than being established in itself.

The model suggests a number of different pathways by which a relatively high and stable homeownership rate is likely to have a positive effect on the vitality of middle neighborhoods, and by extension, how the erosion of homeownership is likely to sap that vitality. As tenure shifts from ownership to rental, under the social and economic conditions affecting those neighborhoods, the neighborhoods are likely to see declines in property improvement and increased mortgage foreclosure and tax delinquency as direct results of the tenure shift. Indirectly, the increased residential instability and reduced investment associated with the erosion of homeownership may in turn lead to reductions in collective efficacy and child outcomes, which in turn may trigger negative changes in crime incidence and property values, both of which are significant destabilizing factors.

I am not suggesting that these changes will necessarily take place. There are far more variables at play than can be suggested by the model, while there is no magic to any particular homeownership rate. However, it is important to stress that the erosion of homeownership in legacy city neighborhoods, particularly since the end of the housing bubble, is not taking place in a social or economic vacuum. It is taking place in the context of a series of powerful demographic and economic trends, all of which are having the effect of placing these neighborhoods increasingly at risk of destabilization. In that context, the erosion of homeownership in legacy cities should be a matter of substantial concern.


As with any complex policy issue, concern does not necessarily offer guidance on how the issue should be addressed. When it comes to the erosion of homeownership, and its effect on middle neighborhoods in legacy cities, this is particularly the case, since any policies to address this particular issue need to be carried out within the context of the highly problematic widespread decline of middle neighborhoods, which imposes significant constraints on what may be fasible.

Figure 5
Conceptual Model of Homeownership and Neighborhood Change

Note: Refers to lower levels of mortgage foreclosure and/or tax delinquency relative to absentee owners.

This is particularly true with respect to what might be seen as the obvious policy solution; namely, to encourage more people to become homeowners in middle neighborhoods. There appear to be severe limitations to what may be possible in this respect. The decline in the number of middle-income households in general, and the number of married-couple child-rearing households not only within the cities but also throughout metropolitan regions, means that the pool from which homebuyers come is a shrinking one. The weak competitive position of many legacy cities in their regions makes them a hard sell for many prospective home-buying households. Although some neighborhoods, with distinct locational, physical or other assets, may, – and should, – become competitive for homebuyers, it is not likely to be an option available for all struggling middle neighborhoods.

A second approach, which is less often discussed but may have a wider potential reach, is how better to retain and engage the neighborhood’s present homeowners, many of whom are not only disengaged but actively fleeing the city for suburban areas.

Slowing their flight and engaging their energies in their neighborhoods are arguably the two most important steps to stabilize these neighborhoods. However, doing so will require some combination of both community-building strategies in the neighborhood– which most probably will depend on the existence of a strong community development corporation (CDC) or other similar entity—and a responsive municipal government capable of improving public services and willing to give its residents a strong role in shaping the destiny of their neighborhoods.

Finally, although this chapter has focused on homeowners, it is important to pay greater attention to the renter population in middle neighborhoods as well as their landlords. Both groups have not received the attention their significant neighborhood role deserves, the former largely ignored and the later often demonized. Both, however, will have a significant impact on their neighborhoods’ future. Creative organizing strategies to engage both tenants and landlords and policy changes that encourage greater stability of tenure for tenants, could be important steps toward greater neighborhood stability, although perhaps not a substitute for homeownership. Moreover, because many tenants eventually do become homeowners, such policies would in all likelihood increase the probability that they buy in the neighborhood, rather than join the flight to the suburbs.


1. It is worth noting, however, that although homeownership may play a more potent ideological role in the United States than elsewhere, when it comes to actual homeownership rates, the United States is roughly in the middle of the pack among developed nations. While the homeownership rate in the United States is higher than that of many European nations like Germany or France, it is much lower than in Italy or Spain, and slightly lower than in other predominately English-speaking countries like the United Kingdom or Canada. In many of these countries, such as Italy, Spain and Israel, homeownership in multifamily housing is much more the norm than in the United States.

2. Cooper 2014

3. Mallach 2011

4. In addition, the problems obtaining reliable data at the neighborhood level are considerable.

5. For obscure historic reasons, the dominant urban neighborhood house form in in a coastal belt including northern New Jersey and most of coastal New England was the two- and three-family house, in which the units were stacked on one another. In Boston, they are known as ‘triple-deckers’. Such houses, while also found elsewhere, make up only a small part of the residential stock in other American cities.

6. This obscures a significant difference between rural and urban housing; in 1900, the non-farm homeownership rate was only 36.4%.

7. Not only is this inherently difficult to measure, but the difficulty is compounded by the effect of homeownership itself; in other words, the process of becoming a homeowner may change the individual’s values and attitudes in significant ways. A fascinating study from Argentina offers strong evidence of those effects (Di Tella et al, 2007).

8. Coulton, Theodos and Turner 2009.

9. Barker and Miller 2009

10. Rohe and Stewart 1996

11. New Jersey landlord-tenant law prohibits eviction except for cause. Tenants are deemed to have indefinite tenure, and unlike most parts of the United States, may not be evicted simply because their lease has expired. Other than for cause, such as non-payment of rent, the only grounds for eviction are that the owner needs the house or apartment for their personal use. Moreover, rent control is permitted at local option by state law, and is widely used. While the average length of tenure for tenants in New Jersey is slightly longer than in the United States as a whole (45% moved in the previous two years, compared to 56%), the difference is roughly proportionate to the difference for homeowners (11 years compared to 13 years), suggesting that the difference is associated with lower in- and out-migration levels for New Jersey, rather than any effect of greater security of tenure on rental stability. However, comparing New Jersey to states with similar migratory profiles, such as Connecticut and Ohio, we find a slight difference (45% in New Jersey compared to 49% in Ohio and 51% in Connecticut), suggesting that the different landlord-tenant regime in New Jersey may have some effect on tenure. If so, it is a very modest one, representing a difference of at most a few months.

12. Mallach 2011.

13. The recent foreclosure crisis has spawned a substantial body of research on the impacts of foreclosure on neighboring properties, which makes a compelling case for its destructive effects. A recent study by Williams, Galster and Verma (2013) is particularly worth noting, in that it found a causal relationship between foreclosure and subsequent decline, as the authors note, “the completed foreclosure indicator was strongly predictive of three other indicators: property crimes, total home purchase loan amounts, and mean home purchase loan amounts (p207).” They characterize foreclosures as an “early warning indicator” of neighborhood change.

14. Ellen at al 2002, Ding and Knapp 2003

15. Coulson, Hwang, and Imai (2002, 2003)

16. Ding and Knapp 2003.

17. Rohe and Stewart 1996.

18. Galster 1987

19. Ioannides 2002

20. Mallach 2014a

21. Mallach 2014b

22. Todd 2010

23. Robinson and Todd 2010, Robinson 2012

24. Green and White 1997

25. Boehm and Schlottmann 1999

26. Rossi and Weber 1996; Diaz-Serrano 2009

27. Rohe and Basolo 1997

28. Manturuk 2012

29. Bowdler, Quercia and Smith 2010, Pollock and Lynch 2009

30. Temkin and Rohe 1998, p. 82. Social capital in their study combined institutional infrastructure and sociocultural milieu, which they define as “a construct that attempts to capture both observable behaviors of neighborhood residents and their unobservable affective sentiments toward the area.” (p. 69)

31. Sampson 2012, p. 27. They defined collective efficacy by (1) constructing an index of social control, in which they asked respondents how they would react (on scale of 1 to 5) to various situations, such as if a fight broke out in front of their house, or they saw children spray-painting graffiti on a nearly building; and (2) constructing a similar index of social cohesion, asking respondents how they felt about statements such as ‘people in this neighborhood can be trusted’. Finding that the two scales correlated very strongly with one another, they combined them to create their measure of collective efficacy.

32. Jacobs 1961. p. 112.

33. (1997), p. 918

34. Ibid. (p. 923)

35. Morenoff, Sampson and Raudenbush 2011

36. DiPasquale and Glaeser 1998, Cheo, Fesselmeyer and Seah 2013

37. Manturuk, Lindblad and Quercia 2010

38. Lindblad, Manturuk and Quercia 2013

39. Lauridsen, Nannerup and Skak 2006

40. Friedrichs and Blasius 2006

41. This is based on the proposition that investor-buyers are significantly less likely to obtain mortgages from HMDA-reporting sources than are homebuyers, particularly first-time homebuyers. This proposition is strongly supported by a 2011 analysis from Campbell/Inside Mortgage Finance, which found that 77% of investor-buyers bought with cash, compared to 26% of ‘move-up’ homebuyers and 10% of first-time homebuyers. See Tracking Real Estate Market Conditions using the Housing Pulse Survey, available at


43. Mallach 2014a

44. Mallach 2015

45. We used on-line databases maintained by the State of New Jersey for all real property records and for real property transactions, singling out what are termed Class 2 (one to four family residential) properties. Purchases by investors were defined as those where (1) the address of the property and the address of the buyer were different; or (2) for transactions where the addresses were the same, where the name of the buyer was clearly not an individual or couple; e.g., “233 Chestnut LLC” or “Flip-That-House, Inc.”

46. It should be possible to calculate the rate of erosion using these data bases by identifying investor vs. owneroccupant sellers as well as buyers. While such an analysis was beyond the scope of the Trenton study or this paper, it would be valuable, and I hope to be able to carry it out in the near future.

47. Regrettably, the ACS data that I used to present citywide data on homeownership erosion in Tables 4A and 4B does not exist in reliable form at the neighborhood (census tract) level. The only data is available from the fiveyear rather than one-year ACS, thus covering a narrower and more uncertain time period, and with a very large margin of error, which is particularly problematic when trying to compare relatively fine-grained changes.

48. The investor share of single family purchases by neighborhood showed very strong correlations (significance level of .99 or greater) with homeownership rate, tax delinquency, violent crime, vacancy and median sales price.

49. Fennell 2009.

50. Haurin and Gill 2002


David Barker and Eric A. Miller, “Homeownership and Child Welfare,” Real Estate Economics 37 (2) (2009): 279-303

Thomas P. Boehm and Alan Schlottmann. “Does Home Ownership by Parents Have an Economic Impact on Their Children?” Journal of Housing Economics 8 (3) (1999): 217-232.

Janice Bowdler, Roberto Quercia, and David A. Smith. The Foreclosure Generation: The Long-Term Impact of The Housing Crisis on Latino Children and Families” (Washington, DC: National Council of La Raza, (2010).

Ryan Cooper. “It’s time to kill the American dream of homeownership” This Week, April 25, 2014

N. Edward Coulson, Seek-Joon Hwang and Susumu Imai. “The Value of Owner-Occupation in Neighborhoods” Journal of Housing Research 13(2) (2002): 153

N. Edward Coulson, Seek-Joon Hwang and Susumu Imai. “The Benefits of Owner-Occupation in Neighborhoods” Journal of Housing Research 14(1) (2003): 21

Claudia Coulton, Brett Theodos and Margery A. Turner. Family Mobility and Neighborhood Change: New Evidence and Implications for Community Initiatives. Washington DC: Urban Institute (2009)

Luis Diaz-Serrano. “Disentangling the housing satisfaction puzzle: Does homeownership really matter?” Journal of Economic Psychology 30(5) (2009): 745-755.

Chengri Ding and Gerrit-Jan Knaap. “Property Values in Inner City Neighborhoods: The Effects of Home Ownership, Housing Investment and Economic Development.” Housing Policy Debate 13(4) (2003), 701-727

Rafael Di Tella, Sebastian Galiani and Ernesto Schargrodsky. “The Formation of Beliefs: Evidence from the Allocation of Land Titles to Squatters” Quarterly Journal of Economics 122 (1) (2007): 209-241.

Ingrid Gould Ellen, Michael Schill, Scott Susin and Amy Ellen Schwartz. “Building Homes, Reviving Neighborhoods: Spillovers from Subsidized Construction of Owner-Occupied Housing in New York City” Journal of Housing Research, 12 (1) (2002): 447-77

Jurgen Friedrichs and Jorg Blasius “Attitudes of Owners and Renters in a Deprived Neighborhood” Paper presented at the ENHR Conference, Ljubljana, Slovakia (2006)

Lee Anne Fennel. The Unbounded Home: Property Values Beyond Property Lines. New Haven: Yale University Press (2009)

George C. Galster. Homeowners and Neighborhood Reinvestment Durham NC: Duke U Press (1987)

Richard K. Green and Michelle J. White. “Measuring the Benefits of Homeowning: Effects on Children,” Journal of Urban Economics 41 (3) (1997): 441-461

Donald R. Haurin and H.Leroy Gill. “The Impact of Transaction Costs and the Expected Length of Stay on Homeownership” Journal of Urban Economics 51 (3) (2002): 563-584

Yannis M. Ioannides. “Residential neighborhood effects.” Regional Science and Urban Economics 32 (2) (2002): 145-165

Jane Jacobs. The Death and Life of Great American Cities New York, NY: Random House (1961)

Jorgen Lauridsen, Niels Nannerup and Morten Skak. “Owner-Occupied Housing and Crime Rates in Denmark” Paper presented at the 2006 ENHR Conference, Ljubljana, Slovakia (2006)

Mark Lindblad, Kim Manturuk and Roberto Quercia. “Sense of Community and Informal Social Control among Lower Income Households: The Role of Homeownership and Collective Efficacy in Reducing Subjective Neighborhood Crime and Disorder.” American Journal of Community Psychology, 51 (1) (2013): 123-139

Alan Mallach. Building Sustainable Ownership: Rethinking Public Policy Toward Lower-Income Homeownership. Philadelphia: Federal Reserve Bank of Philadelphia (2011)

Alan Mallach. “Lessons From Las Vegas: Housing Markets, Neighborhoods, and Distressed Single-Family Property Investors” Housing Policy Debate 24 (4) (2014a): 769-801

Alan Mallach Housing Market Conditions Assessment: City of Flint, Michigan. Flint: Center for Community Progress (2014b)

Alan Mallach. Laying the Foundation for Strong Neighborhoods in Trenton, New Jersey New Brunswick: New Jersey Community Capital and Washington DC: Center for Community Progress (2015)

Kim Manturuk, Mark Lindblad and Robert Quercia. “Friends and Neighbors: Homeownership and Social Capital among Low- to Moderate-Income Families” Journal of Urban Affairs 32 (4) (2010): 471-488

Kim Manturuk. “Urban Homeownership and Mental Health: Mediating Effect of Perceived Sense of Control.” City & Community 11 (4) (2012): 409-430

Jeffrey D. Morenoff, Robert J. Sampson and Stephen W. Raudenbush. “Neighborhood inequality, collective efficacy, and the spatial dynamics of urban violence” Criminology; 39 (3) (2001); 517-558

National Association of Realtors. Social Benefits of Homeownership and Stable Housing. Washington DC: NAR Research Division (2006)

Craig Evan Pollock and Julia Lynch. “Health Status of People Undergoing Foreclosure in the Philadelphia Region” American Journal of Public Health 99 (10) (2009): 1833-1839

Robert D. Putnam, Robert D. “The Prosperous Community: Social Capital and Public Life.” The American Prospect, Spring 1993.

Breck L. Robinson and Richard M. Todd. The Role of Non-Owner-Occupied Homes in the Current Housing and Foreclosure Cycle. Richmond: Federal Reserve Bank of Richmond (2010)

Breck L. Robinson. “The Performance of Non-Owner-Occupied Mortgages During the Housing Crisis”. Economic Quarterly 98 (2) (2012): 111-138

William Rohe and Leslie S. Stewart. “Home Ownership and Neighborhood Stability” Housing Policy Debate 7:1 (1996): 37-81

William Rohe and Victoria Basolo. “Long-Term Effects of Homeownership on the Self-Perceptions and Social Interactions of Low Income Persons” Environment and Behavior 29 (6) (1997): 793-819

Peter H. Rossi and Eleanor Weber. “The Social Benefits of Homeownership: Empirical Evidence from National Surveys” Housing Policy Debate 7 (1) (1996): 1-35

Robert J. Sampson, Stephen W. Raudenbush and Felton Earls. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy” Science 277 (1997): 918-924

Robert J. Sampson. Great American City. Chicago IL: University of Chicago Press (2012)

Kenneth Temkin and William Rohe. “Social Capital and Neighborhood Stability: An Empirical Investigation” Housing Policy Debate 9 (1) (1998): 61-88

Richard M. Todd. Foreclosures on Non-Owner-Occupied Properties in Ohio’s Cuyahoga County: Evidence from Mortgages Originated in 2005–2006. Minneapolis: Federal Reserve Bank of Minneapolis (2010)

Sonya Williams, George Galster and Nandita Verma. “Home Foreclosures as an Early Warning Indicator of Neighborhood Decline” Journal of the American Planning Association 79:3 (2013): 201-210

Writer, scholar, practitioner and advocate, Alan Mallach has been engaged with the challenges of urban revitalization, neighborhood stabilization and housing provision for fifty years. A senior fellow with the Center for Community Progress, he has held a number of public and private sector positions, and currently also teaches in the graduate city planning program at Pratt Institute in New York City. His publications include many books, among them Bringing Buildings Back: From Vacant Properties to Community Assets and A Decent Home: Planning, Building and Preserving Affordable Housing, as well as numerous articles, book chapters and reports. He has a B.A. degree from Yale College, and lives in Roosevelt, New Jersey.

Download the article (pdf, 307.68 kb)

Other articles in this issue

The Middle Neighborhood Movement, 1970-2000

The Case for Intervention in Middle Neighborhoods

Demographics and Characteristics of Middle Neighborhoods in Select Legacy Cities

Is the Urban Middle Neighborhood an Endangered Species? Multiple Challenges and Difficult Answers

Strategies to Improve Middle Neighborhoods

Using Place Branding Strategy to Create Homebuyer Demand for Legacy City Neighborhoods

The Healthy Neighborhoods Program: A Middle Neighborhoods Improvement Strategy

Understanding Middle Market Neighborhoods as Vital Parts of Regional Economies

Rebuilding from Strength as a Strategy to Safeguard Middle Neighborhoods in Detroit: A Philanthropic Perspective

Local Public Policy and Middle Neighborhoods

Preservation in Middle Neighborhoods: Promising Results in Ohio