Temporal and ethnic decompositions of homeownership rates: Synthetic cohorts across five censuses N. Edward Coulson a, * , Maurice Dalton b a Department of Economics, Penn State University, USA b National Bureau of Economic Research, Cambridge, MA, USA article info Article history: Received 8 March 2010 Available online 1 April 2010 JEL classification: R21 Keywords: Homeownership Ethnic differentials Oaxaca decomposition abstract We perform probit-based Oaxaca–Fairlie decompositions of the change in ownership rates for four ethnic groups and three age groups over five censuses, and then construct second- order decompositions of the white/non-white differentials. There is substantial heteroge- neity in how Hispanic, Asian and Black households of various age cohorts converge to and diverge from their white counterparts with respect to homeownership. Ó 2010 Elsevier Inc. All rights reserved. 1. Introduction Homeownership is widely seen as an indicator of social well-being. It is a mechanism for wealth accumulation, and there is some evidence that it creates positive externalities through increased maintenance (Galster, 1983), increased volunteerism and citizenship (DiPasquale and Glaeser, 1999), and improved outcomes for children (Haurin et al., 2002). There is also evidence that such externalities are actually priced in the market for neighboring properties (Coulson et al., 2003) which suggests that these beneficial externalities have value in the marketplace. Therefore, the fact that homeownership probabilities differ systemati- cally across ethnic groups is cause for concern. Aggregate homeownership rates for four ethnic groups for each cen- sus (about which more later) between 1960 and 2000 is provided in Table 1. 1 The four ethnic groups are labeled, despite the inaccuracy that sometimes is inherent in such labels, as white, black, Hispanic and Asian. Three facts are evident: (1) Homeownership rates are rising for all four eth- nic groups over time, with the exception of the early dip in the Hispanic rate, and the overall downward trend from 1980 to 1990 (2) the homeownership rates of black, His- panic and Asian households lag behind that of white house- holds; (3) the relative lagging of these three non-white groups changes over time. In 1960, Hispanics had ownership rates well above those of blacks, but by 1980 the situation was reversed, while Asian households by the same decade had moved ahead of both Hispanics and blacks. The examination of homeownership rate differentials has long been on the research agenda of urban and housing economists and other social scientists. Early concern was with black–white homeownership differentials. Wachter and Megbolugbe (1992), Megbolugbe and Cho (1996), Gyourko and Linneman (1996), Gyourko et al. (1999) Gabriel and Painter (2003), Bostic and Surette (2001), Deng et al. (2003), Painter et al. (2001), Gabriel and Rosenthal (2005), Hilber and Liu (2008) and others have explored various aspects of this differential. The conditional differ- ences between Hispanics and Anglos, and indeed differ- ences within the Hispanic community are examined in Krivo (1995), Coulson (1999), Toussaint-Comeau and Rhine 1051-1377/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jhe.2010.03.001 * Corresponding author. E-mail address: fyj@psu.edu (N.E. Coulson). 1 It is worth noting at the outset that the data in Table 1for Hispanics in the first two decades are, we think, new to this paper, and based on recent modifications to the IPUMS data set as discussed below. Journal of Housing Economics 19 (2010) 155–166 Contents lists available at ScienceDirect Journal of Housing Economics journal homepage: www.elsevier.com/locate/jhec