Defining the American Dream

Measuring the Benefits of Homeowning:

Posted on Tuesday, November 2, 2010

March 5, 1996
In this paper we examine whether homeowning benefits children by testing
whether children of homeowners stay in school longer than children of renters and
whether daughters of homeowners are less likely to have children as teenagers
than daughters of renters. We use both probit models and a bivariate probit
technique which takes account of possible selection bias due to differences between
parents who choose to own versus rent. We find in several data sets that
both effects are statistically significant and quantitatively important}particularly
for low-income households. We also estimate that the dollar benefit per lowincome
household of parents being homeowners rather than renters is at least
$31,000. Q 1997 Academic Press
At least as far back as the 1920s, it has been an article of faith among
policymakers that homeowning is desirable and should be encouraged.
These quotations are illustrative: Herbert Hoover: ‘‘A family that owns its
own home takes pride in it and has a more wholesome, healthful, and
happy atmosphere in which to bring up children’’; Franklin D. Roosevelt:
‘‘A nation of homeowners is unconquerable’’; and Lyndon B. Johnson:
‘‘Owning a home can increase responsibility and stake out a man’s place in
his community.’’ More recent policymakers continue to believe in the value
of homeowning, although they are less specific about its benefits: former
H.U.D. Secretary Jack Kemp: ‘‘Democracy can’t work without the compo-
0094-1190r97 $25.00
Copyright Q 1997 by Academic Press
All rights of reproduction in any form reserved.
nent that goes to the heart of what freedom is all about}the chance to
own a piece of property’’; and the ClintonrGore compaign: ‘‘Homeownership
and decent housing are an essential part of the American Dream.’’1
Policymakers have consistently been willing to back up these sentiments
with public funds: the Office of Management and Budget calculates that
allowing owners to deduct property taxes and mortgage interest payments
from taxable income cost the Federal government $55 billion in foregone
tax revenues in 1993 and allowing homeowners who are over 55 to exclude
$125,000 in capital gains on housing from tax cost an additional $4.4
These statements in effect are claiming that homeowning is a means to a
set of policy ends. Homeowning should be encouraged because owners
take greater responsibility for their families, their communities, and their
country and provide a better environment for their children. Our goal is to
test this view by examining whether children of homeowners behave in
socially more desirable ways than children of renters. In particular we test
whether, controlling for other factors, children of homeowners stay in
school longer than children of renters and are less likely to have children
themselves as teenagers. We use both simple probit models and a bivariate
probit technique which takes account of selection bias due to differences
between parents who choose to own versus rent.
The paper is organized as follows. Section 1 discusses theoretical considerations
and prior research. In Section 2, we investigate probit models of
whether parents’ tenure status affects their children’s success, using three
different data sets. We find that homeowning by parents has a significant
effect on children’s success in all three. In Section 3, we investigate a
bivariate probit model of the stay-in-school decision that attempts to sort
out the effects of homeowning per se from the effects of differences
between parents who choose to own versus rent. In Section 4 we use the
results to calculate a dollar figure for the benefit to children of homeowning
by parents.
1Quotations are taken from Stegman et al. w19x, except for that from the Clintonr Gore
campaign, which comes from ‘‘Clintonr Gore on Affordable Housing for All Americans’’
_campaign statement..
2See Office of Management and Budget w16x, Table 24-1, Part Two, p. 26. An alternate
means of measuring the cost of favorable tax treatment of owner-occupied housing would be
to examine the reduction in Federal tax revenue that results from excluding imputed rent on
owner-occupied housing from taxable income. However, the O.M.B. does not include this in
its computations of tax expenditures. Note that in some years rental housing has also been
treated favorably by the U.S. tax code; but its tax treatment has changed frequently in recent
years. See Gordon et al. w7x.
How might parents’ decisions to own versus rent affect their children’s
behavior? One possibility is that when people own their own homes, they
invest in do-it-yourself skills by learning to do some maintenance jobs
themselves and they also learn financial skills since they must meet the
cost of unexpected home repairs. Homeowners also may learn interpersonal
skills by hiring professionals such as plumbers and roofers or by
pestering City Hall to provide better services. The learning-by-doing model
suggests that the cumulative experience of maintaining a house may cause
homeowners to become better managers. Further, these skills may be
transferable, so that as homeowners learn to better manage their home
environments, they may also take better care of their children. In contrast,
renters are rarely forced to manage their home environments, so they are
less likely to acquire these transferable skills.3
Another difference between homeowners and renters is that homeowners
have a larger financial stake in their neighborhoods, because most of
their wealth is tied up in their residences. Bad behavior by children}their
own or their neighbors’}may reduce the attractiveness of the neighborhood
and threaten the value of their homes. Thus homeowners have a
stronger incentive than renters to monitor their own children and their
neighbors’ children and prevent them from engaging in behavior which
would threaten housing values. This provides an alternate mechanism
through which homeowning may cause better outcomes for children. In
addition, homeowners have higher moving costs and tend to remain in the
same neighborhoods longer than renters. This makes them better at
monitoring and influencing the behavior of children in the neighborhood.
The extra monitoring by adults in neighborhoods dominated by owners is
hypothesized to benefit children.
A contrary view is that children of homeowners do better than children
of renters not because their parents are homeowners, but because their
parents are different. In this view, some parents are more likely both to
buy homes and to bring up successful children, while other parents are
more likely both to rent and to bring up less successful children. If this
view is correct, then homeowning could appear to be a significant determinant
of children’s success because it captures the unmeasured effect of
parents’ personality type rather than because it is important per se. We
consider this possibility in Section 3 below.
3In the learning by doing literature, greater cumulative experience with the production
process reduces production costs and the benefits may accrue either internally to the firm
itself based on its own past production level or externally to other firms based on all firms’
past production levels. See Fudenberg and Tirole w6x.
Most economists probably disagree with the notion that homeowning
can affect behavior: they tend to view whether a household owns or rents
to be merely a financial decision with tax consequences. Research by
economists on owner-occupied housing has instead focused on the fact
that Federal tax treatment of homeowning reduces the per unit price of
housing and gives households an incentive to increase their housing
consumption. Economists have argued that U.S. households consume too
much housing, thereby reducing funds for more productive investments in
plant and equipment.4 They have also examined the distributional impact
of the tax treatment of owner-occupied housing and concluded that the
benefits go disproportionately to upper-income households.5
There is also literature in both sociology and economics on the determinants
of children’s success. Examples include Duncan and Hoffman’s w4x
study of whether black teenage girls have out-of-wedlock children, An et
al.’s w2x study of whether teenage girls have out-of-wedlock children and go
on welfare, Kane’s w13x study of whether blacks graduate from high school
and enter college, Parcel and Menagham’s w17x study of the effect of
parents’ jobs on young children’s cognitive development, and Case and
Katz’ w3x study of neighborhood influences on a variety of young behaviors.
For a review of this literature, see Haveman and Wolfe w10x. With the
exception of the study by Kane, none of these papers considers the
possibility that homeowning by parents might be a determinant of children’s
We start by estimating probit models which explain whether youths
behave in socially desirable ways as a function of whether their families
live in owner-occupied housing and other variables. We focus on 17- or
18-year-old youths because they have been exposed to the maximum
amount of parental influence. But since they are still likely to be living
with their parents, we can determine whether their parents are owners or
renters. We examine whether youths are still in school and whether
daughters have had a child themselves. Three different data sets are
examined: the Panel Study of Income Dynamics _PSID., the Public Use
Microsample of the 1980 Census of Population and Housing _PUMS., and
High School and Beyond _HSB..
4Alm et al. w1x estimated that the benefit]cost ratio for programs to stimulate housing
demand was around 0.6. Also see Hendershott and Shilling w11x, Rosen w18x, and Mills w15x.
5See Rosen w18x and Follain et al. w5x.
PSID Results
The PSID data set consists of children of PSID households who were 17
years old in any of the years 1980]1987. The dependent variable equals
one if youths are still in school or have already graduated from high school
and equals zero if they have dropped out of school. The explanatory
variables are whether the youth’s household lives in owner-occupied housing,
the race of the household head _black equals one., family size, family
income _in thousands of 1982 dollars., a dummy variable for whether the
household head was less than 18 years old when the child was born, three
dummy variables measuring the household head’s educations level: whether
the head graduated from high school, attended college, or graduated from
college _the omitted category is non-high school graduate., whether the
household head is female, whether the household head is divorced, and
whether the household head worked in the past year.6
Table 1, column 1, gives the results.7 _Standard errors are given in
parentheses in all the tables.. The homeowning variable is positive and
significantly different from zero, with a t statistic of 3.25. It thus provides
support for policymakers’ contention that homeowning matters. Of the
other variables, race, family income, whether the head was young when the
child was born, and whether the head worked the past year are also
statistically significant. Since a likelihood ratio test rejects the null hypothesis
that the samples of homeowners and renters come from the same
population, we also run the same probit model separately on the subsamples
of youths whose parents own versus rent. The results are given in
columns 2 and 3 of Table 1. The top portion of Table 2 gives the predicted
probabilities of 17 year olds being in school at different family income
levels, where all variables other than family income are set equal to their
overall mean values. Model 1 gives the predictions from the probit regression
reported in Table 1, column 1, and model 2 gives the predictions from
the probit regressions reported in Table 1, columns 2 and 3. In model 1,
when family income is $10,000 in 1980 dollars _equivalent to $18,000 in
1994 dollars., children of owner-occupiers have a predicted probability of
being in school of 0.91, compared to 0.82 for renters}a 9 percentage
point differential. The differential falls as family income rises, and is 3
percentage points at the average income level. In model 2, when family
OURNAL OF URBAN ECONOMICS 41, 441]461 _1997.

Richard K. Green
School of Business, Uni¨ersity of Wisconsin,
Madison, Wisconsin 53706
Michelle J. White
Department of Economics, Uni¨ersity of Michigan,
Ann Arbor, Michigan 48109
Received January 31, 1996; revised

Supporting Materials


1000 characters maximum Your Name:    

By Category

Recommended Sites