Gentrification and neighbourhood satisfaction: A study of Philadelphia

Abstract

Chinese

Neighbourhood satisfaction is an important facet of life and consideration for policy, as it affects one’s quality of life and well-being, as well as broader residential mobility patterns. While studies have addressed gentrification’s various outcomes, especially residential displacement, few have investigated the relationship between gentrification and neighbourhood satisfaction. Using data from the 2016 Philadelphia Housing and Neighbourhood Survey and mixed-effects logistic regression models, this paper explores the relationship between gentrification and neighbourhood satisfaction, examining how it varies by gentrification type, respondent race, and length of residence. We find that, while gentrification is overall positively associated with neighbourhood satisfaction, its effect on neighbourhood satisfaction is heterogeneous by gentrification type, respondent race, and length of residence. Specifically, when examined by type, only moderate gentrification is positively associated with neighbourhood satisfaction. Hispanic residents and non-Hispanic Black residents are less likely to be satisfied in gentrifying neighbourhoods than in non-gentrifying neighbourhoods. Lastly, gentrification is not significantly associated with neighbourhood satisfaction among long-term residents.

Keywords

gentrification neighbourhood change neighbourhood satisfaction race

Introduction

Neighbourhood satisfaction, defined as ‘residents’ overall evaluation of their neighbourhood environment’ (Hur et al., 2010: 52), has long been an object of interest and inquiry by scholars from a wide range of disciplines and policymakers for several reasons. For one, satisfaction with one’s neighbourhood and residential environment contributes to one’s overall life satisfaction (Lu, 1999; Mohan and Twigg, 2007; Sirgy and Cornwell, 2002) and thereby affects one’s subjective well-being—‘a person’s cognitive and affective evaluations of his or her life as a whole’ (Oishi et al., 2021: 255). Subjective well-being, in turn, is positively associated with objective outcomes such as behaviour, health, educational attainment, and income (De Neve et al., 2013; Diener and Chan, 2011; Lyubomirsky et al., 2005; Oishi et al., 2007). Moreover, residential satisfaction has been found to be a significant predictor of residential mobility patterns, with high satisfaction encouraging residents to stay and promoting residential stability (Lu, 1998; Speare, 1974; Van Assche et al., 2019).

While a large body of literature examines the correlates of neighbourhood satisfaction, less is known about the impact that gentrification may have on residents’ overall perception and evaluation of their neighbourhood. Studies on gentrification have shed light on its negative outcomes for residents, including physical, political, and cultural displacement, increased housing cost burden, and conflicts (Balzarini and Shlay, 2016; Betancur, 2011; Chaskin and Joseph, 2013; Cheshire et al., 2019; Hyra, 2015; Pattillo, 2007; Taylor, 2002), but others have also documented residents’ positive responses to neighbourhood improvements, such as access to more public amenities, goods, and services (Freeman, 2006; Sullivan, 2007; Weil, 2019) and greater perception of neighbourhood collective efficacy (Steinmetz-Wood et al., 2017). Still others have highlighted the context-dependent nature of the relationship between gentrification and neighbourhood satisfaction (Gibbons et al., 2020; Sullivan, 2007), observing differential effects by race.

This paper explores the relationship between gentrification and neighbourhood satisfaction in the context of Philadelphia, using data from the 2016 Philadelphia Housing and Neighbourhood Survey (PHNS) and the census. We find that living in a gentrifying neighbourhood was, overall, positively associated with neighbourhood satisfaction but that, crucially, gentrification’s effect on neighbourhood satisfaction was heterogeneous by gentrification type, respondent race, and length of residence. That is, only moderate gentrification was positively associated with satisfaction, and respondents of racially marginalised groups were less likely to be satisfied if they resided in a gentrifying neighbourhood than if they resided in a non-gentrifying neighbourhood. Lastly, gentrification was not a statistically significant predictor of neighbourhood satisfaction for long-term residents of over 10 years. While not causal, the findings of this paper advance our understanding of gentrification and offer considerations for further research.

Literature review

Theoretical and conceptual background

Scholars have long theorised and conceptualised mechanisms of neighbourhood satisfaction. Campbell et al. (1976) argue that satisfaction with a domain in life, such as the residential environment or neighbourhood, is an outcome of individuals’ perceptions and evaluations of the domain’s objective attributes: personal characteristics influence how one perceives objective attributes and, together with personal standards of comparison, shape how one then evaluates the attributes based on one’s perception, determining the level of satisfaction (see also Marans and Rodgers, 1975). As such, conceptual models that build on Campbell et al. (1976) and Marans and Rodgers (1975) feature three core components that determine neighbourhood satisfaction: (i) objective ‘contextual’ characteristics of the neighbourhood; (ii) objective ‘compositional’ characteristics of the individual; and (iii) subjective perceptions and assessments of the neighbourhood (Amérigo and Aragonés, 1997; Batson and Monnat, 2015; Galster, 1987; Galster and Hesser, 1981; Grogan-Kaylor et al., 2006; Hur et al., 2010; Hur and Nasar, 2014; Jaramillo et al., 2020; Kearney, 2006; Weidemann and Anderson, 1985; also see Cao et al., 2018 for a review).

Studies have identified a long list of determinants of neighbourhood satisfaction across all three components. Regarding objective compositional individual attributes, age, race and ethnicity, income and educational attainment, duration of residence, and tenure have been found to be significant correlates of neighbourhood statisfaction (Basolo and Strong, 2002; Chapman and Lombard, 2006; Galster and Hesser, 1981; Grinstein-Weiss et al., 2011; Lu, 1999; Mohan and Twigg, 2007; Parkes et al., 2002; Permentier et al., 2011; Ren and Folmer, 2017; Rohe and Stegman, 1994; Speare, 1974). The direction and magnitude of the effects of these variables vary within and across studies. Taking tenure as an example, Parkes et al. (2002) find that homeownership is positively associated with neighbourhood satisfaction where homeownership is the dominant form of tenure but negatively associated where it is not; Greif (2015) finds that homeownership may cause owners to respond more sensitively to neighbourhood conditions and local characteristics and amplify both the positive and the negative, leading to divergent effects on neighbourhood satisfaction. These examples highlight the interactions between the individual or household attributes, neighbourhood conditions, and subjective perceptions and assessments.

Regarding the neighbourhood environment, studies have identified various physical and social conditions—both objective characteristics and subjective perceptions and evaluations of the same—that contribute to neighbourhood satisfaction (Amérigo and Aragonés, 1997). Physical factors at the dwelling-unit level include housing type, objective housing conditions and quality, and property values (Basolo and Strong, 2002; Lu, 1999; Mohan and Twigg, 2007; Ren and Folmer, 2017), and self-reported satisfaction with the housing unit is one of the strongest predictors of neighbourhood satisfaction (Lu, 1999; Mohan and Twigg, 2007; Parkes et al., 2002; Permentier et al., 2011). At the neighbourhood level, urban form and density (Hur et al., 2010; Mouratidis, 2018), as well as general appearance of the neighbourhood and both actual and perceived physical upkeep (Ciorici and Dantzler, 2019; Galster and Hesser, 1981; Hur and Nasar, 2014; Parkes et al., 2002) have been found to be significant predictors of satisfaction.

Research shows that social dimensions of the neighbourhood, both objective and subjective, are also crucial for neighbourhood satisfaction. Most resoundingly, lower crime rates and subjective feelings of safety, such as less fear of crime or violence, have both been associated with higher neighbourhood satisfaction (Basolo and Strong, 2002; Chapman and Lombard, 2006; Ciorici and Dantzler, 2019; Hur and Morrow-Jones, 2008; Hur and Nasar, 2014; Parkes et al., 2002; Sampson, 1991). Furthermore, having social ties in the neighbourhood, such as family, close friends, and neighbours, and perceptions of neighbourliness and social cohesion in the neighbourhood more broadly are associated with greater place attachment and neighbourhood satisfaction (Batson and Monnat, 2015; Ciorici and Dantzler, 2019; Dassopoulos et al., 2012; Dassopoulos and Monnat, 2011; Mesch and Manor, 1998; Parkes et al., 2002; Permentier et al., 2011).

Gentrification and neighbourhood satisfaction

Applying the theoretical and conceptual model of neighbourhood satisfaction, one can hypothesise that gentrification’s effects on the objective conditions of the neighbourhood – both physical and social – would affect residents’ subjective perceptions and evaluations of the neighbourhood, which are influenced, in turn, by their objective attributes (e.g. race, tenure, length of residence in the neighbourhood) that determine the nature of their experience of gentrification. Few studies have explored the relationship between gentrification and an explicit measure of neighbourhood satisfaction. McGirr et al. (2015) find that, in a context in which sitting tenants were protected by rent control, longtime residents and newcomers in a gentrifying neighbourhood in Toronto, Canada, expressed similar reactions to neighbourhood change and similar levels of neighbourhood satisfaction. In contrast, Mendoza-Graf et al. (2023) find that residents living in gentrified tracts in two Pittsburgh neighbourhoods experienced smaller increases in their neighbourhood satisfaction than those living in non-gentrifying tracts.

Although research on gentrification and neighbourhood satisfaction is scarce, numerous studies point to gentrification’s effects on various factors associated with neighbourhood satisfaction. In the physical, or non-social, dimension, gentrification and consequent rising housing costs may negatively affect residents. For instance, to avoid displacement out of their gentrifying neighbourhoods, households may be forced to live in lower-quality housing, double up, or pay a higher share of their income in rent (Newman and Wyly, 2006), all of which may lower their neighbourhood satisfaction. However, some residents may positively perceive improvements to amenities and services accompanying gentrification. For example, Freeman (2006) finds in his study of New York’s gentrifying Harlem and Clinton Hill neighbourhoods that residents, in particular the Black middle class, welcomed the improved public amenities and access to better-quality goods and services. Sullivan (2007), from his survey-based research in two gentrifying neighbourhoods in Portland, Oregon, also reports that the majority of the residents were pleased with the way their neighbourhoods had changed. Weil (2019: 1284) finds that, among the ‘layered influences and complexity’ in the narratives of older adults experiencing gentrification in Queens, New York, was an appreciation for ‘reaping the benefits of gentrified infrastructure in their neighbourhood’.

Concerning the social dimension, a large body of literature investigates gentrification as a problem with important consequences for longtime residents (Brown-Saracino, 2017). In particular, qualitative studies document the erosion of community fabric and social cohesion resulting from the influx of newcomers (Betancur, 2011; Pattillo, 2007; Taylor, 2002), as well as the tension and conflicts between gentrifiers and longstanding residents over local services and amenities, lifestyle, social norms, and visions for the community’s future development, resulting in contentious claims to ‘the right to the city’ (Balzarini and Shlay, 2016; Bélanger, 2012; Chaskin and Joseph, 2013; Cheshire et al., 2019; Pattillo, 2007). Further, with newcomers seeking to exert their sensibilities and way of life onto their neighbourhoods, longtime residents, while having managed to avoid physical displacement, may experience political, commercial, and cultural displacement (Hyra, 2015; Monroe Sullivan and Shaw, 2011), loss of neighbourhood character (Zukin, 2009), and loss of sense of place (Shaw and Hagemans, 2015).

These conclusions find some quantitative support: Gibbons et al. (2020) find that community connection is lower in gentrifying neighbourhoods in Philadelphia, and Mendoza-Graf et al. (2023) find that residents in gentrifying neighbourhoods experienced smaller improvements in perceived social cohesion than those in non-gentrifying neighbourhoods. However, as with outcomes in the physical dimension, there are conflicting findings. Steinmetz-Wood et al. (2017), in their study in Montréal, Canada, find that gentrification is positively associated with perception of neighbourhood collective efficacy.

Crucially, studies suggest that race, tenure, and length of residence in neighbourhood may be important factors that shape the nature and extent of the relationship between gentrification and the various outcomes. For example, Sullivan (2007) finds that, while the majority of the residents were pleased with the neighbourhood change, renters and longtime Black residents were less likely to approve of said changes. Gibbons et al. (2020) find that, while gentrification was overall negatively associated with a sense of neighbourhood community, this relationship varied by the resident’s own race and the racial character of neighbourhood change (e.g. whether White- or non-White population increased in the process of gentrification). As such, studies examining gentrification’s influence on neighbourhood satisfaction need to consider the differential effects it may exert on individuals, including how racially stratified housing markets where descrimination persists (Mitchell et al., 2024) shape one’s exposure to, and experience of, gentrification.

Method

Study area: The city of Philadelphia

Philadelphia offers an interesting context in which to study gentrification. It is one of the largest but poorest cities in the US, with nearly 26% of the city’s population living below the poverty line in 2017 (The Pew Charitable Trusts, 2019). In terms of racial composition, according to the 2012–2016 five-year American Community Survey (ACS) data, non-Hispanic Black residents comprise the largest share of the population at 41.6%, while non-Hispanic White residents comprise 35.3%. Furthermore, while Philadelphia remains one of the most racially segregated cities in the US, it has a smaller Black–White homeownership gap than that in the US overall, in part due to the large stock of single-family homes that became accessible to middle- and upper-income Black households between 1960 and 1990 (Whiton et al., 2021). All of these factors together make Philadelphia a compelling place for exploring the relationship between gentrification and neighbourhood satisfaction—in particular, how experiences of gentrification vary by racial group. Scholars have examined gentrification’s effect on residential mobility patterns (Ding et al., 2016; Hwang and Ding, 2020), residents’ credit scores (Ding and Hwang, 2016), neighbourhood community (Gibbons et al., 2020), and perceived levels of housing unaffordability (Gibbons, 2021) in Philadelphia. This study contributes to this body of literature by interrogating neighbourhood satisfaction.

Data

We use novel survey data from the PHNS, which was implemented by the City of Philadelphia (through the Division of Housing and Community Development) and the Philadelphia Housing Authority (PHA) and formed part of the City’s Assessment of Fair Housing in 2016. The survey included a wide variety of questions about housing and neighbourhood issues, including those on the demographic profiles of the respondents, housing type and quality, respondents’ ratings of neighbourhood conditions, and their neighbourhood satisfaction. The survey was available online and in paper, in English and Spanish. The online version was promoted on the city’s website, social media, and traditional media, and the paper surveys were distributed through door-to-door canvassing by community-based organisations and the PHA. Over 5000 residents across every zip code in the city completed the survey. While one of the largest surveys ever administered in Philadelphia on the topic of housing, neighbourhood, and resident sentiment, the PNHS used non-probability sampling, relying on extensive outreach and voluntary participation. As such, the data tended to overrepresent those aged 35 and over, females, Whites, those with a bachelor’s degree or more, and PHA renters at the zip code level and those living in gentrifying zip codes overall. We discuss the implications of this limitation below.

The neighbourhood-level changes and attributes, including gentrification status, are examined for the period from 2000 to 2016, using the 2000 Decennial Census and the 2012–2016 five-year ACS. The census tract is widely regarded an acceptable proxy in quantitative research on neighbourhood change. In this case, however, the only available geographic identifier for the respondents in the PHNS data, and therefore our proxy for neighbourhood, is the zip code. Generally larger than a census tract, the zip code can arguably better capture the spillover effects of gentrification that happens on a smaller geographic scale (e.g. census block group or tract). As census tract boundaries changed between 2000 and 2016, we harmonise them using Brown University’s Longitudinal Tract Data Base crosswalk files (Logan et al., 2014) and subsequently use the ZIP Code Crosswalk Files provided by the US Department of Housing and Urban Development to aggregate tract-level census data up to the zip-code level.

Measuring gentrification

Scholars have proposed various definitions and methodologies for identifying gentrification over the years. In this paper, we adopt the approach devised by Ding et al. (2016) examining gentrification in the Philadelphia context (also used by Ding and Hwang, 2016; Gibbons, 2021; Gibbons et al., 2020; Hwang and Ding, 2020). Consistent with the literature, we conceptualise gentrification as an influx of residents of higher socio-economic status and an increase in housing costs in low-income central-city neighbourhoods. The index, similar to other approaches, first determines the gentrifiability of a neighbourhood: a zip code is eligible to gentrify if its median household income was below the citywide median household income in the starting year, 2000. Zip codes that were not eligible to gentrify are classified ‘non-gentrifiable’.

For a gentrifiable zip code to be ‘gentrifying’, between 2000 and 2016, it must have experienced: (i) a percentage increase in either the median rent or the median home value greater than the citywide median percentage increase; and (ii) an increase in its share of college-educated residents greater than the citywide median increase. We use increase in the share of college-educated residents rather than the median household income in the second condition, as it identifies neighbourhoods with an influx of university students, young professionals, and others who may have low incomes but a higher socio-economic status relative to the incumbent residents (Ding et al., 2016). Further, as educational attainment tends to be more stable than household income, it better identifies influx of outsiders from incumbent upgrading (Clay, 1979; Freeman, 2005). Gentrifiable zip codes that do not meet the above criteria are considered ‘not gentrifying’.

Further, as gentrification is a dynamic process, we categorise the gentrifying zip codes by the intensity of the gentrification process. We adapt Ding et al.’s (2016) approach and classify as: (i) ‘weak gentrification’, those zip codes that are gentrifying but are in the bottom quartile of gentrifying zip codes for rent and housing value in 2016; (ii) ‘moderate gentrification’, those zip codes that are gentrifying and are in the second or third quartile for either rent or housing value in 2016; and (iii) ‘intense gentrification’, those zip codes that are gentrifying and are in the top quartile for rent or housing value in 2016.

Table 1 provides the descriptive statistics of the neighbourhood characteristics by gentrification status. Gentrifying zip codes were located closest to the city centre, followed by non-gentrifying zip codes, then non-gentrifiable zip codes (see Figure 1 below). In 2000, gentrifying zip codes and non-gentrifying zip codes had similar median household incomes and median rents, but gentrifying zip codes had higher median housing values, larger shares of college graduates, non-Hispanic Whites, and renters, and smaller shares of non-Hispanic Blacks and Hispanics on average. Between 2000 and 2016—excluding discussion of attributes that were used to define gentrification—gentrifying zip codes experienced an increase in the share of non-Hispanic Whites and a decrease in non-Hispanic Blacks, in contrast with non-gentrifying zip codes. Lastly, comparison of gentrifying zip codes by intensity reveals that the neighbourhoods varied significantly in their locational, demographic, and housing-related characteristics—in both baseline conditions in 2000 and how they changed in 2016.

Table 1.

Neighbourhood characteristics by gentrification status.

	Non-gentrifiable	Not gentrifying	Gentrifying	Weak	Moderate	Intense
# of zip codes	26	10	10	2	6	2
Mean distance from city hall	9.1 km	5.3 km	1.9 km	2.9 km	2.0 km	0.4 km
Mean neighbourhood characteristics in 2000 (dollar values in 2016 dollars)
Total population	715,116	449,288	348,047	104,696	222,623	20,728
Median household income	$52,651	$33,680	$35,810	$29,533	$36,542	$36,143
Median rent	$763	$561	$583	$521	$593	$714
Median housing value	$121,404	$58,390	$71,030	$46,649	$71,030	$179,049
% Bachelor’s degree	28.4%	9.6%	17.6%	8.6%	13.2%	39.9%
% Non-Hispanic White	62.6%	18.3%	35.9%	3.9%	44.3%	42.6%
% Non-Hispanic Black	27.2%	61.9%	47.9%	92.0%	37.2%	35.8%
% Hispanic	4.2%	15.2%	7.5%	1.5%	9.4%	7.7%
% Owner occupancy	61.6%	57.7%	47.1%	51.0%	53.7%	23.4%
% Change 2000–2016
Total population	5.9%	−2.9%	3.6%	−8.3%	6.5%	33.4%
Median household income	−3.0%	−11.1%	24.3%	−10.6%	21.8%	65.2%
Median rent	9.2%	15.7%	36.8%	21.3%	34.6%	56.4%
Median housing value	55.4%	54.2%	146.0%	134.8%	146.0%	92.5%
% Bachelor’s degree	+8.0	+3.5	+16.0	+8.8	+17.2	+19.7
% Non-Hispanic White	−10.5	−7.4	+5.3	+9.7	+2.2	+10.0
% Non-Hispanic Black	+3.6	+1.3	−11.1	−14.5	−9.0	−14.1
% Hispanic	+4.1	+5.5	+1.6	+2.2	+1.9	+0.2
% Owner occupancy	−4.5	−10.8	−3.8	−10.3	−4.1	+3.8

Source: Authors’ calculations using data from 2000 Census and 2012–2016 ACS.

Figure 1.

(a) Gentrification status (2000–2016) and (b) share of respondents who would recommend their neighbourhood by zip code.

Key variables

Our outcome of interest is overall neighbourhood satisfaction. The question that most closely reflected the concept of neighbourhood satisfaction in the PHNS was: ‘Right now, how likely are you to recommend your neighbourhood to someone else as a good place to live?’ A slight conceptual discrepancy exists between what is asked in the question and how the literature defines neighbourhood satisfaction. By centring ‘someone else’ in the question, it probably has the effect of encouraging the respondent to weight their own desired neighbourhood conditions less and to assess one’s neighbourhood conditions more generally or objectively. We assume that, if one is satisfied with one’s neighbourhood, one is more likely to recommend it to others as a good place to live and take the reported likelihood of recommending one’s neighbourhood as a proxy for neighbourhood satisfaction. Four answer choices were available: ‘Definitely would recommend’, ‘Probably would recommend’, ‘Probably would not recommend’, and ‘Definitely would not recommend’. For simplicity, we combine the first two responses into ‘Recommend’ (=1) and the last two responses into ‘Not recommend’ (=0) to construct a binary dependent variable. Figure 1 shows a map of Philadelphia zip codes by their gentrification status for 2000–2016 and a map of Philadelphia with the share of respondents that would recommend their neighbourhood to others.

Following the conceptual model, we construct and include control variables for each of: (i) individual demographic, socio-economic, and housing characteristics; (ii) individual perception of neighbourhood environment; and (iii) objective neighbourhood characteristics. The literature offers a broad set of correlates of neighbourhood satisfaction. For individual demographic and socio-economic characteristics, we include the respondents’ age, gender, race and ethnicity, educational attainment, and, for housing attributes, we include their length of residence and tenure. For individual perception of neighbourhood environment, we include perceived quality of the housing unit,¹ perceived safety in the neighbourhood,² perceived quality of neighbourhood conditions, amenities and services,³ and perceived level of social cohesion in the neighbourhood.⁴ For objective neighbourhood characteristics, we do not include any zip-code-level variables other than the gentrification index, as our objective is to examine whether gentrifying neighbourhoods are statistically different from non-gentrifying neighbourhoods with regard to neighbourhood satisfaction, not to isolate an independent causal effect of gentrification by controlling for other neighbourhood attributes and processes that may explain or accompany gentrification.

Table 2 shows the descriptive statistics of the variables for the analysis sample, which contains 4558 respondents. Respondents living in non-gentrifiable zip codes had the largest share recommending their neighbourhood to others (80.7%), followed by those living in gentrifying and non-gentrifying zip codes (76.6% and 66.0%, respectively). Among gentrifying zip codes, the share of respondents reporting neighbourhood satisfaction was the highest in zip codes undergoing intense gentrification (83.7%), followed by moderate and weak gentrification (79.3% and 67.3%, respectively). Over 80% of non-Hispanic White respondents were satisfied with their neighbourhoods (83.4%), considerably higher than non-Hispanic Black and Hispanic respondents (68.6% and 68.8%, respectively). Newest entrants to the neighbourhood (0–5 years) had the largest share reporting satisfaction (81.4%), and the shares decreased in ascending order of time in neighbourhood. Similar shares of respondents who owned and privately rented their housing reported that they would recommend their neighbourhood to others (77.4% and 78.8%, respectively). Finally, 82.6% of those who were satisfied with the quality of the housing unit, 91.6% of those who felt safe in the neighborhood, and 88.1% of those who perceived that neighbours relied on each other reported being satisfied with their neighborhood, all significantly higher than their counterparts; those who evaluated their neighbourhood conditions positively were also more likely to recommend their neighborhoods than those who did not. (See appendix for further descriptive statistics of the sample by the gentrification status of zip code).

Table 2.

Descriptive statistics of variables for the analysis sample (n = 4558).

	Variable				% In sample	% ‘Would recommend’ (=1)
Gentrification status of zip code	Non-gentrifiable				39.5	80.7
	Not gentrifying				22.9	66.0
	Gentrifying				37.6	76.6
	Weak gentrification				10.3	67.3
	Moderate gentrification				22.5	79.3
	Intense gentrification				4.7	83.7
Age	18–34 years old				22.7	81.4
	35–64 years old				54.3	72.2
	65 years or older				23.0	78.7
Gender	Male				28.7	81.3
	Female				68.8	73.8
	Other				2.5	66.4
Race & ethnicity	Non-Hispanic White				45.3	83.4
	Non-Hispanic Black				38.9	68.6
	Hispanic				7.5	68.8
	Other				8.4	73.8
Education	No bachelor’s degree				46.5	66.7
Education	Bachelor’s or higher				53.5	83.6
Time in neighbourhood	0–5 years				36.6	81.4
	6–10 years				17.2	77.6
	11–20 years				18.2	73.7
	21 years or longer				28.0	68.7
Tenure	Own				52.9	77.4
	Rent from a private landlord				21.3	78.8
	Rent from PHA				21.4	70.6
	Other				4.3	67.0
Quality of housing	Not satisfied				14.4	35.1
Quality of housing	Satisfied				85.6	82.6
Neighbourhood safety	Does not feel safe				39.9	51.9
Neighbourhood safety	Feels safe				60.1	91.6
Social cohesion	Neighbours do not rely				32.4	50.2
	Neighbours rely on each other				67.6	88.1
Neighbourhood conditions	Min	Mean	Max	SD	Mean for ‘Not recommend’	Mean for ‘Recommend’
	−18.0	0.14	18.0	7.4	−6.1	2.1

Source: Authors’ calculations using data from the PNHS.

Analytical strategy

As the theory and conceptual model of neighbourhood satisfaction specify a multilevel structure—that is, individual-level attributes and perceptions and neighbourhood-level characteristics—and the data accordingly has a nested structure—that is, individuals within zip codes—we employ a series of mixed-effects logistic regression models to estimate the effect of gentrification on neighbourhood satisfaction. The mixed-effects models with random intercepts that vary across zip codes enable us to account for the structure of the data and the potential overestimation of effects. The models can be expressed by the following equations:

\begin{matrix} Level 1 : \ln (\frac{P (Y_{ij} = 1)}{1 - P (Y_{ij} = 1)}) = β_{0 j} + β_{1 j} ATR B_{ij} \\ + β_{2 j} PER C_{ij} \\ + e_{ij} \end{matrix}

Level 2 : β_{0 j} = γ_{00} + γ_{01} GEN T_{j} + u_{0 j}

\begin{matrix} Combined : \ln (\frac{P (Y_{ij} = 1)}{1 - P (Y_{ij} = 1)}) = \\ β_{0} + β_{1} ATR B_{ij} + β_{2} PER C_{ij} \\ + β_{3} GEN T_{j} + u_{0 j} + e_{ij}, \end{matrix}

where $Y_{ij}$ is the outcome variable of whether respondent $i$ would recommend their neighbourhood to others as a good place to live, $ATR B_{ij}$ is the set of individual demographic, socio-economic, and housing-related attributes of respondent $i$ living in zip code $j$ , $PER C_{ij}$ is the set of variables capturing the individual perception of neighbourhood conditions of respondent $i$ living in zip code $j$ , $GEN T_{j}$ is the gentrification status in zip code $j$ , and $e_{ij}$ refers to the $Level 1$ random error and $u_{0 j}$ to the $Level 2$ random error, together comprising the random part of the model.

We address a series of questions sequentially through the models. We exclude non-gentrifiable neighbourhoods from the regression analysis to compare respondents living in gentrifying neighbourhoods with those living in non-gentrifying neighbourhoods. First, we explore the relationship between gentrification and neighbourhood satisfaction. To do so, we start with a base model that regresses neighbourhood satisfaction on individual attributes, perception of neighbourhood conditions, and neighbourhood gentrification status to establish the effect of gentrification on neighbourhood satisfaction in our analysis sample. Next, we look at how this relationship varies by the intensity of the gentrification process. We hypothesise that, the more intense the process, the stronger the effect on neighbourhood satisfaction will be.

Next, we explore if there are differential effects of gentrification on neighbourhood satisfaction depending on the respondent’s race. Given the well-documented history and persistent reality of racial stratification in the housing market (Reina et al., 2021; Rothstein, 2017; Taylor, 2019), we hypothesise that Black or Hispanic respondents would be less likely to be satisfied in gentrifying neighbourhoods due to their heightened vulnerability to its negative impacts, including increased housing costs, threat of displacement, and loss of community. We therefore explore the between-race differences in the effects of gentrification status on neighbourhood satisfaction by introducing interaction terms between race and gentrification status in the models.

Finally, we explore whether the relationship between gentrification and neighbourhood satisfaction varies for long-term residents by focusing on a subset of respondents who have lived in their neighbourhoods for 11 years or longer and who would have experienced much of the gentrification process identified for the period between 2000 and 2016. We are interested in this population for several reasons. Respondents who moved to their neighbourhoods prior to 2005 selected into still gentrifiable neighbourhoods; their subjective perception and assessment of their housing and neighbourhood conditions, including their neighbourhood satisfaction, may vary from those of newer residents, who likely selected into much more clearly gentrifying neighbourhoods and presumably narrowed the gap between their desired and actual neighbourhood conditions (Morris and Winter, 1975). We compare the results for this subset with those of the previous models.

Findings

The model results in Table 3 show that survey respondents living in gentrifying neighbourhoods are more likely to recommend their neighbourhood to others as a good place to live than those living in non-gentrifying neighbourhoods. Controlling for demographic, socio-economic, and housing-related attributes and perceptions of the neighbourhood environment (Model 3), living in a gentrifying neighbourhood increases a respondent’s odds of recommending their neighbourhood by 45.9% ( $e^{0.378} = 1.459$ ).⁵

Table 3.

Model results estimating the effect of gentrification on neighbourhood satisfaction.

	Model 1		Model 2		Model 3
	β	(SE)	β	(SE)	β	(SE)
Gentrification status (Ref: Not gentrifying)
Gentrifying	0.835**	0.264	0.507*	0.22	0.378*	0.172
Demographic & other variables
Non-Hispanic Black (Ref: Non-Hisp. White)			−0.670***	0.144	−0.398*	0.169
Hispanic			−0.427*	0.189	−0.259	0.23
Other race			−0.562**	0.182	−0.086	0.224
18–34 years old (Ref: 35–64 years old)			0.168	0.126	0.338*	0.157
65+ years old			0.526***	0.117	0.217	0.147
Female (Ref: Male)			−0.326**	0.108	−0.181	0.137
Other gender			−0.767**	0.278	−0.505	0.336
Bachelor’s degree			0.417***	0.117	0.254	0.143
6–10 years in neighbourhood (Ref: 0–5 years)			0.004	0.138	−0.012	0.173
11–20 years			−0.227	0.139	−0.460**	0.177
20+ years			−0.275*	0.129	−0.32	0.165
Private renter (Ref: Owner)			−0.042	0.135	0.065	0.171
PHA renter			0.171	0.143	0.061	0.183
Other tenure			−0.222	0.21	0.211	0.266
Perception variables
Housing quality					1.199***	0.138
Neighbourhood conditions					0.117***	0.01
Safety					1.234***	0.124
Social cohesion					1.195***	0.117
Constant	0.507***	0.188	1.136***	0.241	−0.683*	0.296
Random effect variance	0.300		0.186		0.067
Observations	2757		2757		2757
Log Likelihood	−1552.80		−1502.09		−1018.02
Akaike Inf. Crit.	3111.60		3038.18		2078.05
Bayesian Inf. Crit.	3129.30		3138.85		2202.41

p < 0.05; **p < 0.01; ***p < 0.001.

Importantly, though, the strongest predictors of neighbourhood satisfaction are variables pertaining to the respondents’ perceptions of their housing and their neighbourhood environment, as shown in Model 3. This is consistent with previous studies that find subjective variables to be more likely to be significant and/or have greater effect sizes than objective variables (Cao et al., 2018; Jones and Dantzler, 2021; Parkes et al., 2002). Respondents’ perceived nighttime safety in the neighbourhood, satisfaction with the quality of their housing unit, and perceived levels of social cohesion in the neighbourhood each increases a respondent’s odds of recommending their neighbourhood by at least 230%. Respondents’ assessment of neighbourhood conditions, entered as a continuous composite score variable, is also an important predictor variable of neighbourhood satisfaction, with each point increasing the odds of a respondent recommending their neighbourhood by 12.4%.

The coefficients of individual-level attributes show that non-Hispanic Black respondents are significantly less likely to recommend their neighbourhood to others compared to non-Hispanic White respondents, with odds 32.8% lower, even after controlling for respondents’ perception of housing and neighbourhood conditions. Further, respondents who have lived in their neighbourhood for 11–20 years are less likely to report satisfaction than newcomers of 0–5 years, with odds 36.9% lower. Tenure is not a statistically significant predictor of neighbourhood satisfaction controlling for all other variables.

Decomposing the gentrifying zip codes by the level of gentrification—weak, moderate, and intense—we find that the intensity of the gentrification process matters for neighbourhood satisfaction. Table 4 reports the outcomes of Model 4, which replaces the binary variable for gentrification in Models 1, 2, and 3 with a categorical variable of gentrification type. The results suggest that only moderate gentrification is positively associated with neighbourhood satisfaction relative to no gentrification. The odds of a respondent living in a moderately gentrifying neighbourhood recommending their neighbourhood to others is 67.2% higher than those of a respondent living in a non-gentrifying neighbourhood. Despite zip codes undergoing intense gentrification descriptively having the largest share of respondents that would recommend their neighbourhood to others, we do not find support for our initial hypothesis—that is, the more intense the process, the stronger the effect on neighbourhood satisfaction. One possible explanation may be that weak gentrification has yet to materialise observable improvements to amenities and services, while intense gentrification entails changes that are drastic and/or disruptive.

Table 4.

Model results estimating the effect of gentrification by intensity on neighbourhood satisfaction.

	Model 4
	β	(SE)
Gentrification status (Ref: Not gentrifying)
Weak gentrification	0.029	0.228
Moderate gentrification	0.514**	0.179
Intense gentrification	0.342	0.31
Demographic & other variables
Non-Hispanic Black (Ref: Non-Hisp. White)	−0.347*	0.17
Hispanic	−0.246	0.229
Other race	−0.064	0.224
18–34 years old (Ref: 35–64 years old)	0.340*	0.157
65+ years old	0.217	0.147
Female (Ref: Male)	−0.178	0.137
Other gender	−0.487	0.336
Bachelor’s degree	0.264	0.143
6–10 years in neighbourhood (Ref: 0–5 years)	−0.013	0.173
11–20 years	−0.465**	0.177
20+ years	−0.327*	0.165
Private renter (Ref: Owner)	0.071	0.171
PHA renter	0.067	0.182
Other tenure	0.211	0.265
Perception variables
Housing quality	1.195***	0.138
Neighbourhood conditions	0.118***	0.01
Safety	1.230***	0.124
Social cohesion	1.202***	0.117
Constant	−0.708*	0.292
Random effect variance	0.043
Observations	2757
Log Likelihood	−1016.17
Akaike Inf. Crit.	2078.35
Bayesian Inf. Crit.	2214.55

p < 0.05; **p < 0.01; ***p < 0.001.

Next, we examine whether there are any differential effects of gentrification on neighbourhood satisfaction by the respondent’s race and find significant results. Table 5 reports the regression outcomes of Models 5 and 6: Model 5 builds on the base model (Model 3) by introducing interaction terms between race and gentrification status, and Model 6 extends the gentrification type model (Model 4) by introducing interaction terms between race and gentrification type. We find that Hispanic respondents in gentrifying neighbourhoods are less likely to recommend their neighbourhood compared to non-Hispanic White respondents in gentrifying neighbourhoods. The log-odds of a Hispanic respondent in a gentrifying neighbourhood reporting neighbourhood satisfaction are 0.708 less than those of a non-Hispanic White respondent in a gentrifying neighbourhood. Similarly, in weakly gentrifying neighbourhoods, Hispanic respondents and non-Hispanic Black respondents are, respectively, less likely to recommend their neighbourhood to others compared to non-Hispanic White respondents. The log-odds of a Hispanic respondent and a non-Hispanic Black respondent in a weakly gentrifying neighbourhood reporting neighbourhood satisfaction are, respectively, 1.489 and 1.317 less than those of a non-Hispanic White respondent in a weakly gentrifying neighbourhood. While living in a gentrifying neighbourhood is overall positively associated with neighbourhood satisfaction, Hispanic respondents are less likely to recommend their neighbourhood if they reside in one than if they do not, and Hispanic and non-Hispanic Black respondents are less likely to do so if they live in a weakly gentrifying neighbourhood than if they live in a non-gentrifying neighbourhood.

Table 5.

Model results estimating the differential effect of gentrification on neighbourhood satisfaction by respondent race.

	Model 5		Model 6
	β	(SE)	β	(SE)
Gentrification status (Ref: Not gentrifying)
Gentrifying	0.693*	0.29
Hispanic × Gentrifying	−0.900*	0.455
Weak gentrification			0.964	0.494
Moderate gentrification			0.647*	0.3
Intense gentrification			0.591	0.425
Hispanic × Weak gent.			−1.660*	0.79
Non-Hispanic Black × Weak gent.			−1.217*	0.497
Demographic & other variables
Non-Hispanic Black (Ref: Non-Hisp. White)	−0.107	0.262	−0.1	0.26
Hispanic	0.192	0.331	0.171	0.329
Other race	−0.538	0.396	−0.535	0.394
18–34 years old (Ref: 35–64 years old)	0.351*	0.157	0.364*	0.158
65+ years old	0.23	0.147	0.237	0.149
Female (Ref: Male)	−0.189	0.138	−0.189	0.138
Other gender	−0.559	0.339	−0.547	0.346
Bachelor’s degree	0.265	0.144	0.262	0.145
6–10 years in neighbourhood (Ref: 0–5 years)	−0.013	0.173	−0.014	0.175
11–20 years	−0.452*	0.178	−0.471**	0.179
20+ years	−0.299	0.165	−0.319	0.166
Private renter (Ref: Owner)	0.057	0.172	0.036	0.173
PHA renter	0.117	0.185	0.081	0.186
Other tenure	0.193	0.265	0.155	0.266
Perception variables
Housing quality	1.209***	0.139	1.194***	0.139
Neighbourhood conditions	0.117***	0.01	0.117***	0.01
Safety	1.220***	0.124	1.248***	0.126
Social cohesion	1.193***	0.118	1.177***	0.118
Constant	−0.929**	0.342	−0.885**	0.34
Random effect variance	0.075		0.056
Observations	2757		2757
Log Likelihood	−1012.44		−1005.80
Akaike Inf. Crit.	2072.87		2075.60
Bayesian Inf. Crit.	2215.00		2265.10

p < 0.05; **p < 0.01; ***p < 0.001.

Lastly, we test whether the above-examined relationships vary for long-term residents, who have lived in their neighbourhood for over 10 years. Table 6 shows the model results for Model 7, which replicates the base model (Model 3), Model 8, which replicates the race and gentrification interaction model (Model 5), and Model 9, which replicates the race and gentrification type model (Model 6), on the long-term resident subset of the sample. Model 7 indicates that gentrification is not a statistically significant predictor of neighbourhood satisfaction for this subset of respondents. Gentrification by type (model not shown, replicating Model 4) is also not significant. These outcomes suggest that the positive association between gentrification and neighbourhood satisfaction observed in previous models could be driven by newer residents’ response to gentrification. Moreover, the strongest predictor of neighbourhood satisfaction for long-term residents is perception of social cohesion—whether neighbours can count on each other for help; it has a larger effect size than both perceived housing unit quality and perceived safety in the neighbourhood, which is not the case in the previous all-sample models.

Table 6.

Model results estimating the effect of gentrification on neighbourhood satisfaction for long-term residents.

	Model 7		Model 8		Model 9
	β	(SE)	β	(SE)	β	(SE)
Gentrification status (Ref: Not gentrifying)
Gentrifying	0.202	0.244	0.499	0.449
Hispanic × Gentrifying			−0.897	0.682
Non-Hispanic Black × Gentrifying			−0.345	0.483
Other × Gentrifying			0.284	0.721
Weak gentrification					1.007	0.858
Moderate gentrification					0.151	0.466
Intense gentrification					1.34	0.778
Hispanic × Intense gentrification					−2.742*	1.133
Demographic & other variables
Non-Hispanic Black (Ref: Non-Hisp. White)	−0.501*	0.245	−0.295	0.382	−0.276	0.379
Hispanic	−0.419	0.335	−0.018	0.47	−0.043	0.469
Other race	−0.043	0.344	−0.222	0.581	−0.196	0.581
18–34 years old (Ref: 35–64 years old)	−0.134	0.32	−0.149	0.321	−0.146	0.329
65+ years old	0.445*	0.189	0.451*	0.189	0.454*	0.193
Female (Ref: Male)	−0.233	0.198	−0.234	0.198	−0.23	0.203
Other gender	−0.7	0.471	−0.726	0.473	−0.82	0.507
Bachelor’s degree	0.007	0.191	0.016	0.191	−0.005	0.198
Private renter (Ref: Owner)	0.276	0.284	0.254	0.286	0.278	0.292
PHA renter	0.002	0.239	0.015	0.243	−0.056	0.247
Other tenure	0.369	0.415	0.362	0.414	0.314	0.43
Perception variables
Housing quality	1.013***	0.203	1.022***	0.203	1.031***	0.206
Neighbourhood conditions	0.101***	0.014	0.102***	0.014	0.104***	0.014
Safety	1.240***	0.175	1.228***	0.176	1.277***	0.18
Social cohesion	1.327***	0.168	1.316***	0.168	1.326***	0.172
Constant	−0.818*	0.392	−1.005*	0.466	−1.008*	0.465
Random effect variance	0.133		0.154		0.120
Observations	1189		1189		1189
Log Likelihood	−485.906		−484.581		−471.241
Akaike Inf. Crit.	1007.81		1011.16		1000.48
Bayesian Inf. Crit.	1099.27		1117.86		1147.83

p < 0.05; **p < 0.01; ***p < 0.001.

Examining Models 8 and 9 with interaction terms for race and gentrification status, we find that race and gentrification interaction terms are not significant in Model 8, but the interaction term between Hispanic and intense gentrification is significant in Model 9. Among long-term residents, the log-odds of a Hispanic respondent living in an intensely gentrifying neighbourhood reporting neighbourhood satisfaction are 2.785 less than the same of a non-Hispanic White respondent living in an intensely gentrifying neighbourhood. We discuss the implications of these findings below.

Discussion and conclusion

This paper explored the relationship between gentrification and neighbourhood satisfaction through a study of Philadelphia. Living in a gentrifying neighbourhood was, overall, positively associated with the likelihood of recommending one’s neighbourhood to others as a good place to live, controlling for several objective and subjective factors. However, when decomposing the gentrification status by the intensity of the gentrification process, we found that only moderate gentrification was positively associated with neighbourhood satisfaction, whereas intense gentrification had no significant relationship. Furthermore, respondents of racially marginalised groups—non-Hispanic Black respondents and, particularly, Hispanic respondents—were less likely to be satisfied if they resided in a gentrifying neighbourhood than if they resided in a non-gentrifying neighbourhood.

There are several possible reasons for the differential effects of gentrification on the neighbourhood satisfaction of Hispanic and Black respondents. The first is their potentially heightened financial vulnerability to gentrification’s negative impacts because of wealth inequality and racial stratification and discrimination in the housing market (Massey and Denton, 1993; Oh and Yinger, 2015; Reina et al., 2021; Rothstein, 2017; Taylor, 2019). The second is the potential loss of community in gentrifying neighbourhoods. As reported in Table 1, gentrifying zip codes experienced significant losses in the share of the Black population. Such changes in racial composition were starker in weakly gentrifying and intensely gentrifying zip codes, which were not positively associated with neighbourhood satisfaction and, in fact, negatively associated with neighbourhood satisfaction for Hispanic and Black respondents.

We found that gentrification was not a significant predictor of neighbourhood satisfaction for long-term residents and that, for a particular subset of long-term residents (i.e. long-term Hispanic residents living in zip codes of intense gentrification), it was a negative predictor. The results also highlighted the importance of perceived social cohesion in the neighbourhood for long-term residents in our sample. As such, not only are long-term residents distinguishable from new entrants in their response to gentrification more specifically, but they may also differ in their criteria and processes of evaluation of their residential environment more broadly. Scholars have long theorised and observed that long-term residents may become less satisfied due to life-cycle changes over time, rendering their neighbourhoods inadequate (Speare, 1974). However, our findings suggest that they may derive relatively greater satisfaction from factors such as social cohesion and relations in the neighbourhood, which require time investments to materialise. Future research could directly test this interpretation.

Our study has several limitations. First, using cross-sectional data, we are unable to draw conclusions on causality or the mechanisms driving the relationship between gentrification and neighbourhood satisfaction. Second, the PNHS is not a true representative sample. As aforementioned, even though it has a large sample size, it relied on non-probability sampling and, therefore, a number of variables have distributions that statistically differ from those reported in the census at the zip code level (e.g. underrepresentation of Blacks and Hispanics). Perhaps more importantly, it also could suffer from selection bias. That is, those who have been most adversely impacted by gentrification—that is, those who have been physically displaced from their homes and neighbourhoods due to rising housing cost burden—may not be captured in the sample as they are no longer present in the gentrifying neighbourhoods (Atkinson, 2000). Moreover, there is a significant, unmeasured type of displacement, which Marcuse (1986) terms exclusionary displacement. As Newman and Wyly (2006: 27) summarise: ‘Neighbourhoods become off-limits, forcing low-income residents to look to lower-cost neighbourhoods for housing’. Indeed, Hwang and Ding (2020) find that gentrification reduces the pool of neighbourhoods accessible to financially disadvantaged residents, leading them to relocate to disadvantaged tracts that are not gentrifying. Taken together, the limitations suggest that the associations uncovered in this paper between gentrification and neighbourhood satisfaction, particularly for Black and Hispanic respondents, may be a plausible lower bound and requires further investigation. In any case, a true representative sample is unknown, and findings from this study—which leverages a large and unique, albeit imperfect, dataset from the city—should be generalised with caution.

Our study presents several policy implications. The racialised relationship between gentrification and neighbourhood satisfaction calls for more significant efforts to further fair housing and to increase enforcement of fair housing principles and policies so that residents of racially marginalised groups can exercise choice in the housing market and attain quality housing and neighbourhoods. Furthermore, policymakers should reduce potential risks of displacement by preserving and expanding access to affordable homeownership options, especially in gentrifying neighbourhoods, as well as boosting the affordable rental supply and enacting stronger renter protections. Moreover, given the importance of subjective perceptions of housing quality, safety, and neighbourhood conditions for neighbourhood satisfaction, investments in the rehabilitation of aged or substandard housing, infrastructure, and amenities, especially in neighbourhoods that have faced systemic disinvestment, should be a priority. That is, in tandem with policy responses to gentrification, policymakers should pursue equitable investment in non-gentrifying neighbourhoods so that more households, particularly those that have previously been marginalised, experience the positive gains of investment, coupled with access to affordable and decent housing.

Footnotes

Appendix

Appendix.

Descriptive statistics of the sample by gentrification status of zip code (n = 4558).

		Non-gentrifiable	Not gentrifying	Gentrifying	Weak	Moderate	Intense
	# of respondents	1801	1045	1712	471	1026	215
Age	% 18–34 years old	19.6	17.7	29.0	26.3	30.2	29.3
	% 35–64 years old	57.4	56.3	49.9	49.0	49.4	54.4
	% 65 years or older	23.0	26.0	21.0	24.6	20.4	16.3
Gender	% Male	31.5	25.9	27.3	21.0	28.4	36.3
	% Female	66.1	71.7	69.9	68.9	60.5	66.1
	% Other	2.4	2.8	2.4	2.8	2.7	3.3
Race & ethnicity	% Non-Hispanic White	61.6	17.0	45.3	21.9	52.8	60.5
	% Non-Hispanic Black	25.4	61.7	39.2	66.2	32.0	14.4
	% Hispanic	5.2	14.0	5.9	4.0	6.4	7.4
	% Other	7.8	7.3	9.6	7.9	8.8	17.7
Education	% Bachelor’s or higher	64.5	34.9	53.3	37.6	56.8	71.2
Time in neighborhood	% 0–5 years	32.5	32.7	43.4	38.9	44.3	48.8
	% 6–10 years	16.8	14.7	19.2	20.6	18.4	20.0
	% 11–20 years	21.7	16.9	15.2	17.6	14.3	14.4
	% 21 years or longer	29.1	35.6	22.1	22.9	22.9	16.7
Tenure	% Own	66.5	41.0	45.9	29.7	53.3	45.6
	% Rent from a private landlord	20.4	21.9	22.0	19.3	20.2	36.3
	% Rent from PHA	9.7	29.6	28.9	48.2	23.0	14.4
	% Other	3.4	7.6	3.3	2.8	3.5	3.7
Quality of housing	% Satisfied	90.9	79.7	83.6	77.5	85.4	88.4
Neighborhood safety	% Feels safe	71.3	46.5	56.7	47.6	59.8	61.9
Social cohesion	% Neighbours rely on each other	71.8	62.3	66.4	64.1	67.9	64.2
Neighborhood conditions	Mean composite score	2.46	−1.57	−1.25	−1.69	−1.38	0.32

Source: Authors’ calculations using data from the PNHS.

Acknowledgements

The authors are grateful for the helpful feedback and suggestions of Keith Wardrip, attendees of a seminar held at the Federal Reserve Bank of Philadelphia, and the anonymous reviewers at this journal. We also thank Claudia Aiken for her data support and feedback on the paper. The views and findings in this paper are those of the authors.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Yeonhwa Lee

Vincent J. Reina

Notes

References

Amérigo

Aragonés

(1997) A theoretical and methodological approach to the study of residential satisfaction. Journal of Environmental Psychology 17(1): 47–57.

Atkinson

(2000) Measuring gentrification and displacement in Greater London. Urban Studies 37(1): 149–165.

Balzarini

Shlay

(2016) Gentrification and the right to the city: Community conflict and casinos. Journal of Urban Affairs 38(4): 503–517.

Basolo

Strong

(2002) Understanding the neighborhood: From residents’ perceptions and needs to action. Housing Policy Debate 13(1): 83–105.

Batson

Monnat

(2015) Distress in the desert: Neighborhood disorder, resident satisfaction, and quality of life during the las Vegas foreclosure crisis. Urban Affairs Review 51(2): 205–238.

Bélanger

(2012) The meaning of the built environment during gentrification in Canada. Journal of Housing and the Built Environment 27(1): 31–47.

Betancur

(2011) Gentrification and community fabric in Chicago. Urban Studies 48(2): 383–406.

Brown-Saracino

(2017) Explicating divided approaches to gentrification and growing income inequality. Annual Review of Sociology 43(1): 515–539.

Campbell

Converse

Rogers

(1976) The Quality of American Life: Perceptions, Evaluations, and Satisfactions. New York, NY: Russell Sage Foundation.

10.

Cao

Yuan

(2018) Examining built environmental correlates of neighborhood satisfaction: A focus on analysis approaches. Journal of Planning Literature 33(4): 419–432.

11.

Chapman

Lombard

(2006) Determinants of neighborhood satisfaction in fee-based gated and nongated communities. Urban Affairs Review 41(6): 769–799.

12.

Chaskin

Joseph

(2013) ‘Positive’ gentrification, social control and the ‘right to the city’ in mixed-income communities: Uses and expectations of space and place. International Journal of Urban and Regional Research 37(2): 480–502.

13.

Cheshire

Fitzgerald

Liu

(2019) Neighbourhood change and neighbour complaints: How gentrification and densification influence the prevalence of problems between neighbours. Urban Studies 56(6): 1093–1112.

14.

Ciorici

Dantzler

(2019) Neighborhood satisfaction: A study of a low-income urban community. Urban Affairs Review 55(6): 1702–1730.

15.

Clay

(1979) Neighborhood Renewal: Middle-Class Resettlement and Incumbent Upgrading in American Neighborhoods. Lexington, MA: Lexington Books.

16.

Dassopoulos

Batson

Futrell

, et al. (2012) Neighborhood connections, physical disorder, and neighborhood satisfaction in Las Vegas. Urban Affairs Review 48(4): 571–600.

17.

Dassopoulos

Monnat

(2011) Do perceptions of social cohesion, social support, and social control mediate the effects of local community participation on neighborhood satisfaction? Environment and Behavior 43(4): 546–565.

18.

De Neve

J-E

Diener

Tay

, et al. (2013) The objective benefits of subjective well-being. In: Helliwell

Layard

Sachs

(eds) World Happiness Report 2013. New York, NY: UN Sustainable Development Solutions Network, pp.1-35.

19.

Diener

Chan

(2011) Happy people live longer: Subjective well-being contributes to health and longevity. Applied Psychology. Health and Well-Being 3(1): 1–43.

20.

Ding

Hwang

(2016) The consequences of gentrification: A focus on residents’ financial health in Philadelphia. Cityscape 18(3): 27–55.

21.

Ding

Hwang

Divringi

(2016) Gentrification and residential mobility in Philadelphia. Regional Science and Urban Economics 61: 38–51.

22.

Freeman

(2005) Displacement or succession?: Residential mobility in gentrifying neighborhoods. Urban Affairs Review 40(4): 463–491.

23.

Freeman

(2006) There Goes the Hood: Views of Gentrification From the Ground Up. Philadelphia, PA: Temple University Press.

24.

Galster

(1987) Identifying the correlates of dwelling satisfaction: An empirical critique. Environment and Behavior 19(5): 539–568.

25.

Galster

Hesser

(1981) Residential satisfaction: Compositional and contextual correlates. Environment and Behavior 13(6): 735–758.

26.

Gibbons

(2021) Measuring gentrification’s association with perceived housing unaffordability: A Philadelphia case study. Housing Policy Debate 31(2): 306–325.

27.

Gibbons

Barton

Reling

(2020) Do gentrifying neighbourhoods have less community? Evidence from Philadelphia. Urban Studies 57(6): 1143–1163.

28.

Greif

(2015) The intersection of homeownership, race and neighbourhood context: Implications for neighbourhood satisfaction. Urban Studies 52(1): 50–70.

29.

Grinstein-Weiss

Yeo

Anacker

, et al. (2011) Homeownership and neighborhood satisfaction among low- and moderate-income households. Journal of Urban Affairs 33(3): 247–265.

30.

Grogan-Kaylor

Woolley

Mowbray

, et al. (2006) Predictors of neighborhood satisfaction. Journal of Community Practice 14(4): 27–50.

31.

Hur

Morrow-Jones

(2008) Factors that influence residents’ satisfaction with neighborhoods. Environment and Behavior 40(5): 619–635.

32.

Hur

Nasar

(2014) Physical upkeep, perceived upkeep, fear of crime and neighborhood satisfaction. Journal of Environmental Psychology 38: 186–194.

33.

Hur

Nasar

Chun

(2010) Neighborhood satisfaction, physical and perceived naturalness and openness. Journal of Environmental Psychology 30(1): 52–59.

34.

Hwang

Ding

(2020) Unequal displacement: Gentrification, racial stratification, and residential destinations in Philadelphia. American Journal of Sociology 126(2): 354–406.

35.

Hyra

(2015) The back-to-the-city movement: Neighbourhood redevelopment and processes of political and cultural displacement. Urban Studies 52(10): 1753–1773.

36.

Jaramillo

Rohe

Webb

(2020) Neighborhood opportunity and satisfaction among housing choice voucher recipients: A subjective well-being perspective. Housing Policy Debate 30(5): 703–717.

37.

Jones

Dantzler

(2021) Neighbourhood perceptions and residential mobility. Urban Studies 58(9): 1792–1810.

38.

Kearney

(2006) Residential development patterns and neighborhood satisfaction: Impacts of density and nearby nature. Environment and Behavior 38(1): 112–139.

39.

Logan

Stults

(2014) Interpolating U.S. decennial census tract data from as early as 1970 to 2010: A longitudinal tract database. Professional Geographer 66(3): 412–420.

40.

(1998) Analyzing migration decisionmaking: Relationships between residential satisfaction, mobility intentions, and moving behavior. Environment and Planning A 30(8): 1473–1495.

41.

(1999) Determinants of residential satisfaction: Ordered logit vs. regression models. Growth and Change 30(2): 264–287.

42.

Lyubomirsky

King

Diener

(2005) The benefits of frequent positive affect: Does happiness lead to success? Psychological Bulletin 131(6): 803–855.

43.

Marans

Rodgers

(1975) Toward an understanding of community satisfaction. In: Hawley

Rock

(eds) Metropolitan America in Contemporary Perspective. New York, NY: John Wiley & Sons, pp.299–352.

44.

Marcuse

(1986) Abandonment, gentrification, and displacement: The linkages in New York City. In: Smith

Williams

(eds) Gentrification and the City. Abingdon, Oxon: Routledge, pp.153–177.

45.

Massey

Denton

(1993) American Apartheid: Segregation and the Making of the Underclass. Cambridge, MA: Harvard University Press.

46.

McGirr

Skaburskis

Donegani

(2015) Expectations, preferences and satisfaction levels among new and long-term residents in a gentrifying Toronto neighbourhood. Urban Studies 52(1): 3–19.

47.

Mendoza-Graf

Collins

Dastidar

, et al. (2023) Changes in psychosocial wellbeing over a five-year period in two predominantly Black Pittsburgh neighbourhoods: A comparison between gentrifying and non-gentrifying census tracts. Urban Studies 60(6): 1139–1157.

48.

Mesch

Manor

(1998) Social ties, environmental perception, and local attachment. Environment and Behavior 30(4): 504–519.

49.

Mitchell

Edlebi

Meier

, et al. (2024) Decades of Disinvestment: Historic Redlining and Mortgage Lending Since 1981. 28 May. National Community Reinvestment Coalition. Available at: https://ncrc.org/decades-of-disinvestment/ (accessed 31 May 2024).

50.

Mohan

Twigg

(2007) Sense of place, quality of life and local socio-economic context: Evidence from the survey of English housing, 2002/03. Urban Studies 44(10): 2029–2045.

51.

Monroe Sullivan

Shaw

(2011) Retail gentrification and race: The case of Alberta Street in Portland, Oregon. Urban Affairs Review 47(3): 413–432.

52.

Morris

Winter

(1975) A theory of family housing adjustment. Journal of Marriage and Family 37(1): 79–88.

53.

Mouratidis

(2018) Is compact city livable? The impact of compact versus sprawled neighbourhoods on neighbourhood satisfaction. Urban Studies 55(11): 2408–2430.

54.

Newman

Wyly

(2006) The right to stay put, revisited: Gentrification and resistance to displacement in New York City. Urban Studies 43(1): 23–57.

55.

Yinger

(2015) What have we learned from paired testing in housing markets? Cityscape 17(3): 15–60.

56.

Oishi

Diener

Lucas

(2007) The optimum level of well-being: Can people be too happy? Perspectives on Psychological Science 2(4): 346–360.

57.

Oishi

Diener

Lucas

, et al. (2021) Subjective well-being: The science of happiness and life satisfaction. In: Snyder

Lopez

Edwards

(eds) The Oxford Handbook of Positive Psychology, 3rd edn. New York, NY: Oxford University Press, pp.254–264.

58.

Parkes

Kearns

Atkinson

(2002) What makes people dissatisfied with their neighbourhoods? Urban Studies 39(13): 2413–2438.

59.

Pattillo

(2007) Black on the Block: The Politics of Race and Class in the City. Chicago, IL: University of Chicago Press.

60.

Permentier

Bolt

van Ham

(2011) Determinants of neighbourhood satisfaction and perception of neighbourhood reputation. Urban Studies 48(5): 977–996.

61.

Reina

Pritchett

Wachter

, et al. (2021) Perspectives on Fair Housing. Philadelphia, PA: University of Pennsylvania Press.

62.

Ren

Folmer

(2017) Determinants of residential satisfaction in urban China: A multigroup structural equation analysis. Urban Studies 54(6): 1407–1425.

63.

Rohe

Stegman

(1994) The effects of homeownership: On the self-esteem, perceived control and life satisfaction of low-income people. Journal of the American Planning Association 60(2): 173–184.

64.

Rothstein

(2017) The Color of Law: A Forgotten History of How Our Government Segregated America. New York, NY: Liveright Publishing Corporation.

65.

Sampson

(1991) Linking the micro- and macrolevel dimensions of community social organization. Social Forces 70(1): 43–64.

66.

Shaw

Hagemans

(2015) ‘Gentrification without displacement’ and the consequent loss of place: The effects of class transition on low-income residents of secure housing in gentrifying areas. International Journal of Urban and Regional Research 39(2): 323–341.

67.

Sirgy

Cornwell

(2002) How neighborhood features affect quality of life. Social Indicators Research 59(1): 79–114.

68.

Speare

(1974) Residential satisfaction as an intervening variable in residential mobility. Demography 11(2): 173–188.

69.

Steinmetz-Wood

Wasfi

Parker

, et al. (2017) Is gentrification all bad? Positive association between gentrification and individual’s perceived neighborhood collective efficacy in Montreal, Canada. International Journal of Health Geographics 16(1): 24.

70.

Sullivan

(2007) Reassessing gentrification: Measuring residents’ opinions using survey data. Urban Affairs Review 42(4): 583–592.

71.

Taylor

K-Y

(2019) Race for Profit: How Banks and the Real Estate Industry Undermined Black Homeownership. Chapel Hill, NC: University of North Carolina Press.

72.

Taylor

(2002) Harlem Between Heaven and Hell. Minneapolis, MN: University of Minnesota Press.

73.

The Pew Charitable Trusts (2019) The State of Philadelphians Living in Poverty, 2019. April. Philadelphia, PA: The Pew Charitable Trusts. Available at: https://www.pewtrusts.org/-/media/assets/2019/05/state_of_poverty.pdf (accessed 1 October 2022).

74.

Van Assche

Haesevoets

Roets

(2019) Local norms and moving intentions: The mediating role of neighborhood satisfaction. Journal of Environmental Psychology 63: 19–25.

75.

Weidemann

Anderson

(1985) A conceptual framework for residential satisfaction. In: Altman

Werner

(eds) Home Environments. Human Behavior and Environment. Boston, MA: Springer US, pp.153–182.

76.

Weil

(2019) Relationship to place for older adults in a New York City neighborhood undergoing gentrification: A discourse analysis. City and Community 18(4): 1267–1286.

77.

Whiton

Singleton

Ding

(2021) What’s Behind the Racial Homeownership Gap in Philadelphia?December. Philadelphia, PA: Federal Reserve Bank of Philadelphia. Available at: https://www.philadelphiafed.org/-/media/frbp/assets/community-development/briefs/cdro-brief-homeownership2-final.pdf (accessed 25 March 2022).

78.

Zukin

(2009) Naked City: The Death and Life of Authentic Urban Places. New York, NY: Oxford University Press.