Poverty, affluence, and income inequality: neighborhood economic structure and its implications for health
Introduction
Since the publication of Wilson's The Truly Disadvantaged (1987), there has been a resurgence of scholarly interest in neighborhood structural effects on residents’ physical and mental well-being. This contextually oriented health research has been spurred by increasing interest in macro-social influences on individual health status, the availability of advanced statistical techniques for fitting multi-level regression models (Pickett & Pearl, 2001), and a more general concern with the limitations of exclusively individual-level research designs. In this area of research, economic conditions are the most frequently examined structural factors thought to be relevant for health status over and above individual characteristics. Three aspects of economic context have been discussed in the literature—the level of concentrated poverty, the prevalence of affluent families, and the degree of income inequality within the neighborhood.
Initially, health-related contextual effects research focused on the relationship between spatially concentrated deprivation and health. Findings from this research supported the hypothesis that residence in a poverty neighborhood area has negative effects for a range of health-related outcomes including all cause mortality (Haan, Kaplan, & Camacho, 1987), self-rated health and medical conditions (Luo & Wen, 2002), disease incidence (Barr, Diez-Roux, Knirsch, & Pablos-Méndez, 2001), incidence of severe childhood injury, (Durkin, Davidson, Kuhn, O’Connor, & Barlow, 1994), depression (Yen & Kaplan, 1999), intimate partner violence (Cunradi, Caetano, Clark, & Schafer, 2000), health behaviors such as physical activity level (Yen & Kaplan, 1998) and alcohol-related problems (Jones-Webb, Snowden, Herd, Short, & Hannan, 1997). On the other end of the scale, spatially concentrated affluence per se has seldom been directly examined in health research (but see Browning, Cagney, & Wen, forthcoming; Browning & Cagney, 2002a). The relative absence of research linking neighborhood affluence with health has persisted despite initial evidence that the presence of high- and low-income residents may exert unique effects on health. For instance, Browning and Cagney (2002a) found that affluence was positively associated with fair or poor self-rated health in a large urban sample. Cagney et al. (2002) found that this effect held for older adults as well. Affluence has also been demonstrated to affect neighborhood-level social conditions that may have implications for individual well-being, including health status (Brooks-Gunn, Duncan, Klebanov, & Sealand, 1993). Sampson and colleagues found that concentrated affluence was a more consistent predictor of intergenerational closure and reciprocal exchange than concentrated disadvantage (Sampson, Morenoff, & Earls, 1999). Some studies have used composite scores that combine indicators of neighborhood poverty and affluence (Ross & Mirowsky, 2001). Although composite indices are statistically efficient and parsimonious (Pickett & Pearl, 2001), they may mask the relative contributions of their component measures.
A more recent hypothesis focuses on the shape of the income distribution within a community or a society as a predictor of its overall health level, and presumably individual health status as well (Kaplan, Pamuk, Lynch, Cohen, & Balfour, 1996; Kennedy, Kawachi, & Prothrow-Stith, 1996; Lynch, Smith, Kaplan, & House, 2000; Marmot & Wilkinson, 2001; Wilkinson, 1996). This hypothesis suggests that what really matters to health is relative income; i.e., people's relative socioeconomic standing in relation to others. A substantial body of ecological evidence notwithstanding, this hypothesis remains largely untested for individual health, especially with respect to the impact of inequality within smaller spatial aggregations such as the neighborhood.
Implicit in the discussion of economic structure effects on health are two approaches: (1) an additive perspective on the link between neighborhood socioeconomic status (SES) and health that hypothesizes unique effects of poverty, affluence, and inequality (“poverty is bad, poverty and inequality are worse”); and (2) a perspective that places dimensions of economic structure in competition, arguing that effects of one dimension may be spurious due to association with another (“poverty isn’t the problem, it's inequality”). Drawing on extant theory, we describe and test an additive approach to the effects of economic conditions on health placing particular emphasis on the impact of concentrated affluence on health. Our analyses examine the relative influence of the three measures of socioeconomic context on self-rated health after controlling for individual socio-demographic and socioeconomic characteristics as well as aggregated educational attainment in the neighborhood. We also explore the mechanisms linking neighborhood economic conditions and individual health. The results of our analyses call into question implicit reliance on an additive effects model and point to the need for more careful theoretical and empirical work identifying dimensions of neighborhood SES that are relevant to health, the pathways through which economic structure influences are channeled, and the appropriate level of analysis for the investigation of contextual effects. To our knowledge, no previous research has comparatively evaluated the effects of neighborhood concentrated affluence, concentrated poverty and income inequality in relation to individual health.
Section snippets
Previous studies
Both Wilson (1987) and Massey (1996) have documented the geographic concentration of poverty in large American cities. In The Truly Disadvantaged (1987), Wilson pointed out that the number of people living in poverty areas (defined as census tracts with poverty rates of at least 20 percent) rose by 40 percent in the five largest US cities between 1970 and 1980. Over the same period, the number of people living in high-poverty areas (those with poverty rates of at least 40 percent) grew by 69
Physical, service and social resources
A number of potential mechanisms underlying the impact of neighborhood-level economic context on health have been proposed. Macintyre, Maciver, and Sooman (1993), conceptualized socio-environmental influences on health as falling into five broad types: physical features of the environment shared by all residents in a locality (e.g., quality of air and water); the availability of healthy/unhealthy environments at home, work, and play (e.g., decent housing, secure and non-hazardous employment);
Data
Four data sources were used to explore the above hypotheses. They were: (1) the 1990 Decennial Census; (2) the 1994–95 Project on Human Development in Chicago Neighborhoods-Community Survey (PHDCN-CS); (3) the 1991–2000 Metropolitan Chicago Information Center Metro Survey (MCIC-MS); and (4) Homicides in Chicago, 1965–1995 (Homicides 65–95). Both neighborhood and individual measures were derived from these data. A final multi-level sample was constructed linking these multi-level measures by
Results
Table 4 reports the results of a series of hierarchical ordinal logistic models, assessing the relative importance of affluence, poverty and income inequality at the neighborhood level to individual self-rated health. Model 1 reports the results of a model predicting self-rated health based on socioeconomic and demographic background factors. Consistent with expectations and previous research, age, African-American race, Latino ethnicity, cigarette smoking and high blood pressure are all
Discussion
The purpose of this work was two-fold. First, we attempted to verify that contextual economic conditions exerted influence on individual self-rated health independent of individual SES, and to assess the relative importance of three contextual measures, affluence, poverty and income inequality, for individual self-reported health. Second, we explored mechanisms operating at the neighborhood level that may help to explain the influence of structural economic conditions on health. Responding to
Acknowledgements
We wish to thank Robert Sampson, Felton Earls, and members of the Project on Human Development in Chicago neighborhoods for generously providing access to the Community Survey data. We also thank the Metropolitan Chicago Information Center for providing access to the Metro Survey data. We thank Robert Sampson, Diane Lauderdale and Kate Pickett for their helpful comments. Direct correspondence to Ming Wen, University of Chicago, 5841 South Maryland Avenue, MC 2007, Chicago, IL 60637.
References (72)
- et al.
Neighborhood poverty as a predictor of intimate partner violence among White, Black, and Hispanic couples in the United StatesA multilevel analysis
Annals of Epidemiology
(2000) - et al.
Individuals and their ecologiesAnalyzing the geography of chronic illness within a multilevel modelling framework
Health & Place
(1995) - et al.
Deprivation and mortalityThe implications of spatial autocorrelation for health resources allocation
Social Science & Medicine
(2001) - et al.
Place effects on healthHow can we conceptualize, operationalize and measure them
Social Science & Medicine
(2002) Social capital and health (plus wealth, income inequality and regional health governance)
Social Science & Medicine
(2002)- et al.
Access to medical care in the USRealized and potential
Medical Care
(1978) - et al.
Exploring dimensions of access to medical care
Health Services Research
(1983) - et al.
Social capital
(2000) - et al.
Neighborhood poverty and the resurgence of tuberculosis in New York City, 1984–1992
American Journal of Public Health
(2001) - et al.
Do neighborhoods influence child and adolescent development
American Journal of Sociology
(1993)