Hanratty, Maria H., Sara A. McLanahan,
and Becky Pettit, “The Impact
of the Los Angeles Moving to Opportunity Program on Residential Mobility,
Neighborhood Characteristics, and Early Child and Parent Outcomes”, Bendheim-Thoman Center for Research on Child Wellbeing
Working Paper Number 98-18, Princeton University, April 1998.
This study provides an evaluation of the early impacts of Los
Angeles site of the Moving to Opportunity
program (MTO). It finds that both Experimental and Section 8 groups moved to
neighborhoods with much higher socioeconomic levels than the Control group.
While on average, Experimental and Section 8 neighborhoods were often similar,
the Experimental group was both more likely to move to low-poverty
neighborhoods (58% vs. 1%) and more likely to remain in a high-poverty
neighborhoods (32% vs. 17%) than the Section 8 group. Both Experimental and
Section 8 groups achieved substantial reductions in neighborhoods crime rates
and substantial increased in perceived neighborhood safety levels relative to
the Control group. In addition, parents in both treatment groups report increased
in hours and earnings, increases in utilization of center based child care, and
reductions in hospital emergency care. With respect to social capital, the
evidence is mixed. Parents in both treatment groups report reductions in church
activity and are somewhat less likely to have friends and family in their
neighborhood than the control group. However, these parents are no less likely
to be involved in their children’s activities, and their children are just as
likely to have friends in the neighborhood.
- Does a program which offers a
combination of financial incentives and supportive services increase the
number of housing voucher recipients that move to (and remain in) low
- What is the impact of
neighborhood characteristics on the well-being of parents and children?
- MTO baseline survey,
including most of the initial family characteristics of the sample of all
MTO families who entered the Los Angeles MTO program between 3/10/1995 and
- A follow-up phone survey
conducted between September and December of 1996, and complemented by
in-person visits to those households not reached by phone. A second phone
survey (without site visits) of families that entered the program between
4/24/96 and 12/18/96 was conducted between August and October of 1997. The
combined response rate for both surveys was 81% (285 or 354), which
reflects a rate of 88% for the Experimental group, 65% for the Section 8
group, and 81% for the Control group. The second survey sample had a
higher proportion of blacks (54% vs. 32%), never-married households (58%
vs. 34%), and households that were employed at the baseline (41% vs. 23%).
In the few areas in which results are sensitive to these differences, the
authors discuss how the results would vary if the second sample were
- Experimental families
reported that supportive services were critical to relocating to non-poor
areas, but many also reported that they would not have moved to their
current neighborhood had they not been constrained by the low-poverty
- Overall, 67% of Experimental
families moved, compared to 85% of Section 8 and 27% of Control group
families. The differences between Experimental and Section 8 in this
category have decreased over time.
- At the time of the follow-up
survey, 58% of the Experimental group lived in low poverty neighborhoods,
compared to only 1% of both the Section 8 and Control groups. By contrast,
32% of MTO families remained in very high poverty neighborhoods, compared
to 17% of Section 8 and 80% of the Control group.
- Both the Experimental and
the Section 8 groups reported marked reductions in neighborhood crime
rates and improvements in perceived neighborhood safety.
- Factors correlated with
housing mobility for the combined sample of Section 8 and Experimental
families include: Age -- Younger household heads were more likely to move;
and Victimization status -- households with a member who had been
assaulted or beaten in the last six months were more likely to move.
However, a key factor related to mobility for the Experimental
sample was marital status -- household heads never married were more
likely to move.
- Employment and community
resources: both treatment respondents in both treatment groups worked more
hours and had higher earnings than did control group members. Experimental
families were more likely to use center-based child care and Head Start
programs (which are generally considered to be superior to other child
care) and were more likely to receive medical care at a doctor’s office.
All experimental group families were less likely to receive medical care
in an emergency room.
- Social capital: this
evidence is mixed. Treatment groups were just as likely as the control
group to be involved in their children’s school activities and to belong
to "other organizations". But treatment groups also had fewer
friends and family in their neighborhood, and reported slightly lower
levels of church membership and attendance.
outcomes: Children in the two treatment groups were doing well, despite
the fact that many of them were having to adjust to a new neighborhood.
They were not isolated and they appeared to be making new friends in their
new neighborhoods. This finding holds for adolescents as well as
elementary age children, which is surprising, given that younger children
are expected to adjust more easily to a move than adolescents. Although
there was some evidence that children in the two treatment groups may have
been experiencing a small deficit in total activities, as compared with
control children, the differences were small and likely to be temporary.
The fact that parents were establishing contact with other parents in the
neighborhood is also a good sign and suggests that residential mobility
does not necessarily result in a major loss of social capital, at least
for this population of families.
Pettit, Becky, Sara A. McLanahan, and Maria Hanratty, “Moving to Opportunity: Benefits and Hidden
Costs,” Bendheim-Thoman Center for Research on Child
Wellbeing," Princeton University, Bendheim-Thoman
Center for Research on Child Wellbeing Working Paper 98-11, February 2000.
This paper uses data from the Moving to Opportunity (MTO) Program
in Los Angeles, California to examine the effects of a
housing voucher program on neighborhood resources. We examine neighborhood quality
as well as neighborhood connections, and we examine the potential
trade-off between these two types of resources. We find that moving is
associated with dramatic improvements in neighborhood quality, as measured by
crime rates, poverty rates, employment rates, and the prevalence of college
graduates and female-headed families. We also find that moving has surprisingly
few negative effects on neighborhood connections. There is some evidence that
moving to a middle class neighborhood reduces participation in after-school
activities among young children and African American children.
- Do housing vouchers improve
- Does moving reduce social
- Does moving to a middle
class neighborhood reduce social capital more than moving per se?
- Is moving to a middle class
neighborhood more difficult for African American families than for other
From four sources:
- In-person baseline survey of
- A follow-up telephone survey
designed and conducted approximately one year after the program began
(between September 1996 and September 1997). The authors indicate that
they use all four data sources to examine the experiences of 284 adults
and 331 children in 231 families, but they do not state which of these
participated in phone survey, or the phone survey response rate.
- 1990 census data on
neighborhood quality; and
- local crime statistics from
the Los Angeles Police Department and the Federal Bureau of Investigation.
The authors created indices of neighborhood quality, parents’ neighborhood
connections, children’s neighborhood connections, and residential mobility.
Multiple regression was used to estimate the effects of observed residential
mobility on each outcome, controlling for a set of individual characteristics.
The authors also used two-stage least squares to estimate the effects of predicted
residential mobility on each outcome, controlling again for a set of individual
- Families given the
opportunity to move through vouchers experienced improved neighborhood
quality, through major reductions in neighborhood murder rates, poverty
rates, female headship rates, and major improvement in employment rates
and rates of college graduation.
- Moving did not reduce
social capital, at least as measured by parent connections with other
parents and teachers, or their participation in work, school, or religious
- In general, families who
moved to non-poor neighborhoods were equally likely to be connected with
neighbors and local institutions as were parents who moved to poorer
neighborhoods or who did not move at all. The exception to this was that
moving to non-poor neighborhoods seems to have reduced the number of
after-school activities in which children participated. The reasons for
this are not clear.
- Although most of the
evidence suggests that there were no important race/ethnic differences in
the effects of mobility on neighborhood connections, moving up does
appear to reduce a black child’s chances of participating in multiple
after-school activities more than it reduces a non-black child’s chances.
Pettit, Becky, “Moving and
Children's Social Connections: The Critical Importance of Context,” Bendheim-Thoman Center for Research on Child Wellbeing,
Working Paper 98-04, Princeton University, March 2000.
Moving during childhood is associated with declines in educational
achievement, educational attainment, and early adult occupational outcomes.
Coleman (1988, 1990) and others have argued that the negative effects of moving
for children may be due to the loss of social capital in the short-term after
moving. There have been few studies directly examining the consequences of
moving on the social connections of children, and the evidence on the
relationship is mixed. This research uses qualitative data from an experimental
housing relocation program to examine what hurts and what helps the formation
of social connections after moving. This research suggests that the impact of
moving on children, and on indicators of social capital in particular, is
influenced by neighborhood context and by family financial resources. Future
studies assessing the impact of moving on children need to pay closer attention
to the factors that influence where, when, and why families move.
- What are the direct
short-term effects of moving on children's social connections?
- How does neighborhood
context influence the hypothesized relationship between moving and social
connections important for children?
Data are drawn from three sources:
- In-depth interviews
conducted with heads of household in 27 families enrolled in the MTO
program in Los Angeles
County in the spring
of 1996, and drawn from four census tracts. The Experimental sample was
stratified by random assignment and destination neighborhood. Due to small
sample size, efforts were made to contact all members of the Section
8-only group in the four census tracts. A small number of control
participants were also contacted for interviews. 27 of 28 heads of
household contacted agreed to be interviewed, with Spanish speaking
participants slightly underrepresented among the Section 8-only
- Data on neighborhood
quality, drawn from the 1990 census.
- Local crime statistics from the
Los Angeles Police Department (1995) and the Federal Bureau of
- Parents and their children
are more likely to make social connections in middle class neighborhoods,
which are more often perceived to be safe, than in poor neighborhoods.
- Financial capital leads to
social capital, and as such, poor families are handicapped when social
connections and activities in low-poverty neighborhoods require money that
parents do not have.
- The data suggests that
relationship between moving and social capital may be influenced by the
factors that influence who moves, where, and when (e.g. whether moves are
effected by family disruptions or factors that may impact social capital
Matulef, Mark and Manuel Pastor, “Moving to Opportunity (MTO)
Demonstration for Fair Housing: Los Angeles Demonstration Site Interim Outcomes
of Housing Search and Counseling Strategies: Early Lessons for Experimental
Design and Implementation,” December 26, 1999.
This study combines quantitative and qualitative measures to evaluate the
efficacy of the MTO program on two levels: 1) the effectiveness of the program
in successfully placing families in low-poverty areas; and 2) the value of the
Los Angeles MTO program as an experiment with results that can be generalized
to other populations. The program as implemented in Los Angeles is found to yield some positive
results: most prominently, the majority of the MTO participants were issued
vouchers and were able to lease-up with their subsidies in neighborhoods with
less poverty, higher employment rates, higher rates of high school graduation,
and more ethnically diverse populations. However, issues such as the
unfamiliarity of the Non-Profit Organizations (NPOs)
with the target neighborhoods hampered the progress of the program initially.
Unique characteristics of the Los
Angeles context, as well as idiosyncratic
implementation of the program by the NPOs, advise
against blithe cross-site comparisons.
- How effective was MTO in
placing families in low-poverty areas?
- What is the value of the Los
Angeles MTO program as an experiment with results that can be generalized
to populations beyond the program participants?
DATA AND DESIGN
The authors combined quantitative data with interviews and observations.
Specifically, data sources included:
- MTO initial baseline survey
and monitoring data.
- Group and individual
interviews with housing authority and non-profit organization NPO) staff.
- Facilitated discussion with
the entire MTO administrative team and counseling staff.
- Participant focus groups
(one in English and one in Spanish)
- Intensive participant
- The Multifamily Tenant
Characteristics System (MTCS), which consists of
data from public housing agencies as reported to HUD.
- 1990 US Census data.
This study has four primary components:
- Examination of the baseline
data, comparing sample families with public housing residents in the
nation, California, and Los Angeles.
- Examination of the MTO
Experimental group support process, so as to compare the design for
"minimal but stark" differences between the treatment and
control groups in terms of the support provided to MTO families in
- Spatial analysis of MTO
families at baseline, core move, and relocation period.
- Evaluation of interviews
with staff and MTO families regarding attitudes toward their new
neighborhoods and concerns regarding future plans.
- Broadly speaking, the
program was implemented successfully (from the perspective of local
program officials). That is, most families assigned to the MTO
Experimental or Section 8 groups received a voucher and were able to find
suitable housing and lease-up under the subsidy.
- The NPOs
performed inconsistently. They were not familiar with the services or
landlords in the target neighborhoods, and hence, there was a significant
learning curve for NPO staff. Services provided
to MTO households appeared to vary significantly between households, while
the division of labor between the two NPOs
- Anxiety about their moves to
low-poverty areas as well as search and moving costs were significant for
the MTO Experimental families.
- There was no systematic
investigation of violations of fair housing or discrimination laws, which
may be a significant barrier to housing mobility.
- The goals of MTO may
conflict with other HUD programs which encourage geographic stability.
Brennan, Brian. “Los Angeles.”
Moving To Opportunity Research.
Created August 30, 2000. Last Modified August 5, 2002. http://www.nber.org/mtopublic/LA.htm.
Return to MTO Main Page