There's no place like home: the relationship of nonstandard employment and home ownership over the 1990s

American Journal of Economics and Sociology, The, Oct, 2004 by Barbara A. Wiens-Tuers

If changing job structures are impacting the ability of some households to own a home, accumulate assets, and increase social capital, this could have important long-term consequences not only for individual households, but their communities as well. The goal of this paper, then, is to look at the relationship of types of employment and rates of home ownership. The first section of the paper presents summary statistics from the 1979 National Longitudinal Survey of Youth on home ownership over the period 1994 to 1998. The next section uses logistic regression analysis to test for the impact of the type of employment on the probability of home ownership. A discussion of the results of the regression and some concluding remarks make up the final sections of the paper.

II

Is There a Correlation Between the Type of Employment and Home Ownership?

RESEARCHERS USING CPS AND AHS DATA find similar trends in home ownership by income group and ethnicity over the 1990s expansion period. In general, the two data sets show increased rates of home ownership for nonwhites and for lower income levels, although both note that the rate of growth is slower than for whites and higher income levels (Bostic and Surette 2000; Orr and Peach 1999). These trends broadly parallel that of data coming out of the 1979 National Longitudinal Survey of Youth (NLSY79). The NLSY79 is a longitudinal data set comprised of surveys of 12,686 males and females that were between the ages of 14 and 21 in 1979, Following the survey participants for the next decade and a half (1979-1994), the NLSY79 contains detailed employment histories as well as demographic and socioeconomic measures. The participants in the survey represent the tail end of the baby boom who entered, moved through the labor force, and made homing decisions during the same time period. Starting in 1994, the survey began inquiries into the respondents' type of employment arrangements.

Table 2 lists home ownership rates for comparable years from CPS. AHS, and the NLSY79. The results show little significant difference between the rates of home ownership calculated from the data sets for the years 1997 and 1998. What is of note, however, is the difficulty of comparing results by race and ethnicity across different data sets since "white" and "others" are defined differently for the different data sets. In the NLSY79 data set, "others" represents respondents from the initial 1979 survey who did not identify themselves as either black or Hispanic in answer to the ethnicity question or whose race was coded "white" or "other" (Center for Human Resource Research 1999). The category "others" may be construed to represent predominately white households and other nonblack and non-Hispanic households for the NLSY79.

Table 3 breaks down the NLSY79 data for 1994, 1996, and 1998 for all households by ethnicity and by income group. In 1994 the cohort ranged in age from 29 to 36 years of age. In 1996 the age range is from 31 to 38 years of age, and in 1998 the ages range from 33 to 40. The age of the cohort must be kept in mind when looking at the rates of home ownership, since almost all studies of rates of home ownership indicate increasing age is a significant demographic variable in its prediction. The category for all participants of the survey shows an increase in the rate of home ownership from 55.4% in 1994 to 64.0% in 1998. Rates of home ownership for black and Hispanic households increased over the period but remain well below the rate of home ownership for "others." Home ownership by income level represents what would be expected, in that the rate of home ownership goes up as household income increases, although the lowest income group showed a large increase in home ownership over the period.