Publication DateAugust 2002
Author(s)Michael Hubbard, Walter R. Davis
This working paper presents the center’s theoretical approach to the study of the social impacts of homeownership through the first long-term study of a group of low-income homeowners and a comparison panel of renters.This working paper presents our theoretical approach to the study of the social impacts of homeownership. Previous studies of these social impacts have primarily focused on basic differences in economic and social outcomes or psychological status between owners and renters without providing insight into how homeownership brings about these outcomes.
In an extensive review of literature on the social impact of homeownership, Rohe, McCarthy and Van Zandt (2000) identified two major shortcomings of existing research. Their first conclusion was that future research needs to do a better job of identifying processes or mechanisms through which homeownership influences the different social variables of interest.
Second, they concluded that future research needs to do a better job of addressing the self-selection bias inherent in research on the impacts of homeownership. That is, these studies have been unable to isolate the effects of homeownership, making it impossible to know if the attitudes, behaviors, and social outcomes of owners are the result of homeownership or if people who hold such attitudes or are likely to experience such social outcomes are more likely to become homeowners.
We propose to address these shortcomings of previous research in three ways. First, we use current social-psychological theories of rational action, the Theory of Planned Behavior and its extension the Model of Goal-Directed Behavior, to investigate the psychological antecedents of behavior and how these interact with the social environment. By focusing on a small set of behaviors through this theoretical lens, we will be able to observe if and how decision-making processes differ for owners and renters. This theoretical approach is potentially applicable to any number of behaviors, but we plan to focus on four: neighborhood involvement, savings behavior, parenting behavior (for respondents with children), and intention to buy a home (for renters).
Second, there have been few longitudinal studies of the social impact of homeownership, and those that have been published involved general-purpose datasets such as the Panel Survey of Income Dynamics or census data. HUD’s Moving to Opportunity research program has followed several panels of public housing families, but its focus has not been on homeownership. To date, there have been no longitudinal surveys designed with the specific intent of following new homeowners or renters.
In our surveys, we will measure how the attitudes, behavior, and social capital of new homeowners and
comparable renters change over time. We will be able to investigate how neighborhood conditions, employment stability, family stability, and other factors interact with homeownership status.
By combining the information from these two surveys, we will be better able to isolate the independent effects of homeownership relative to existing studies on the impacts of homeownership. By following a sample of owners and renters over five years and measuring changes in the same constructs, we will be able to determine if the changes are greater for homeowners and whether neighborhood conditions and other factors impact those changes in the same way.While only an experimental design with random assignment could truly isolate the impacts of homeownership, the CAP/renters study will allow us to more thoroughly isolate these impacts than previous studies have done.
By virtue of the design of the CAP/renter study, a longitudinal survey of low- and moderate-income renters, we will be able to contribute to the literature in a third way. An underlying and so far unaddressed question for all studies of the social impacts of homeownership is whether people with a given set of characteristics self-select into homeownership. That is, does homeownership cause certain beneficial social outcomes and attitudes, or are people with those attitudes more likely to become homeowners and more likely to experience those outcomes regardless of tenure?
With the renter panel, we will be able to address this selectivity question by following renters who become homeowners. This will allow us to discover whether renters who become homeowners during the life of our panel are different from those who do not in terms of their attitudes, perceived social norms, social capital, and involvement in their neighborhoods.
Furthermore, we will be able to compare the pre-purchase attitudes, norms, and so on of renters who become owners to the post-purchase attitudes, norms, and so on of the owners, thereby assessing whether these supposed “outcomes” of homeownership are in fact antecedents. Although we will contribute to the literature on the social impacts of homeownership in three significant ways, the primary purpose of this paper is to focus on the first: outlining our social-psychological theoretical approach to how homeownership impacts social outcomes.
This paper is organized in three sections. The first section briefly reviews other theoretical approaches and their empirical support. The second section presents our theoretical approach: the Theory of Planned Behavior and its extension, the Model of Goal-Directed Behavior. The third section summarizes our approach to the measurement of social impacts in the homeowner and renter questionnaires. It lays out the areas in which we will use our theoretical approach as well as other areas of interest.