J Popul Econ (2002) 15:527–548

Child care, women’s employment, and child outcomes

Jane Waldfogel
Columbia University School of Social Work, 622 W. 113th Street, New York, NY 10025, USA
(Fax: þ1-212-854-2975; e-mail: jw205@columbia.edu)

Received: 9 December 1999/Accepted: 20 September 2000

Abstract. This paper reviews the evidence on the impact of child care and
maternal employment in the pre-school years on child outcomes. This topic
has long been of interest to economists, developmental psychologists, and
scholars from other disciplines, and has been the focus of increased attention
in recent years, as research has provided additional evidence about the pro-
cesses of development in the earliest days, weeks, and years of life.1
In this paper, I review the evidence on two broad sets of questions: what
we know about the potential benefits of early intervention child care programs,
and what we know about the e¤ects (whether positive or negative) of maternal
employment and child care in the first years of life. The evidence reviewed in
this paper suggests that we now know a good deal about both sets of questions.
But, this review also suggests that there are important gaps in our knowledge
that future work by economists could fruitfully address.

JEL classification: D1, J1, J2

Key words: Child care, female employment, child development, early inter-
vention

1. Background
There is a large literature in economics on the e¤ects of child care costs and
other characteristics on the employment and child care decisions of women

An earlier version of this paper was completed while I was a visitor at the Centre for Analysis
of Social Exclusion at the London School of Economics. I am grateful to colleagues there, and
to John Ermisch and two anonymous referees, for many helpful comments. I would also like to
gratefully acknowledge support from the William T. Grant Foundation and the National Institute
for Child Health and Development. Responsible editor: John F. Ermisch
528 J. Waldfogel

with pre-school age children.2 These studies have sought to establish how sen-
sitive mothers’ decisions about working and using child care are to the cost of
child care and to other characteristics of care such as quality and availability.
Implicit in this literature is the notion that when parents make such deci-
sions, they take into account the impact that their decisions will have on out-
comes for their children. All else equal, we would expect parents to seek to
make child care and employment arrangements that they believe will produce
the best outcomes for their children. Yet, many children are in care that is of
poor or mediocre quality (Helburn (ed.) 1995), and parents’ decisions about
child care are not always responsive to what developmental psychologists
view as indicators of quality – structural features such as smaller group sizes,
higher sta¤/child ratios, and higher levels of provider training. David Blau
and Alison Hagy (1998) o¤er three possible explanations for this paradox.3
First, parents may not agree with child development professionals about the
definition of quality child care. This could come about because parents do not
know what constitutes high-quality child care from a developmental perspec-
tive, or because they value other characteristics of care (such as familiarity, re-
liability, or convenience) more highly (Sonnenstein 1991; Helburn (ed.) 1995).
A second explanation is that parents may actually agree with developmental
psychologists about what constitutes quality care, but may not be responsive to
the indicators that are typically used to assess the quality of child care because
these indicators may not be good measures for it. The evidence on this point
is mixed: some studies have found that indicators such as smaller group size,
higher sta¤/child ratios, and higher levels of provider training are positively
associated with quality (Ruopp et al. 1979; Phillips 1987; Mocan et al. 1995),
but others have not (Blau 1997). A third explanation is that parents’ decision-
making is constrained by the supply of care available; that is, parents may
value quality and may agree that care arrangements with smaller group sizes,
higher sta¤/child ratios, and more provider training o¤er better quality, but
may be unable to find care of this type in their community. The evidence on
this point is mixed as well: Blau and Hagy (1998) find little evidence of supply
constraints in their data, but other studies have reported shortages of care
of particular types in particular communities (Hayes et al. 1990). This latter
point raises the issue of a¤ordability, which may be a fourth explanation for
the mismatch between the quality of the arrangements we think parents would
want for their children and the quality of the arrangements they actually pur-
chase. Although Blau and Hagy (1998) argue that if parents are not willing to
pay the market price for high quality care arrangements, no supply constraint
exists, other child care analysts (see, for instance, Gormley 1995) argue that
there is an a¤ordability problem in the child care market in that parents with
limited incomes can not a¤ord to pay the market price for high quality care
arrangements.
Whatever the reason for the disconnect between parents’ child care deci-
sions and what developmentalists see as indicators of quality (and probably
there is some truth in each of the explanations considered above), the end re-
sult is that some children end up in child care arrangements that are of less
than optimal (or even adequate) quality. A similar situation may arise when it
comes to parental decisions about employment when children are very young.
That is, when mothers of pre-school age children make decisions about work-
ing in the labor market, we would expect them to seek to make decisions that
lead to the best outcomes for their particular child(ren). However, in making
Child care, women’s employment, and child outcomes 529

these decisions, parents will be weighing potential benefits as well as costs.

Maternal employment may produce benefits for children through its impact
on family income, at the same time that it may produce costs for children
through its impact on parents’ availability. Moreover, parents’ decisions about
whether to work, what hours to work, and in what type of jobs may be con-
strained by a number of factors, such as their other potential sources of in-
come, the employment options in their community, and their level of knowl-
edge and their beliefs about the likely impact of their employment on their
children’s development.
The upshot of this discussion is that we can not predict a priori the impact
of parents’ decisions about child care and employment in the pre-school years
on outcomes for their children. Although well-meaning parents will seek to
make the best possible decisions for their children, there may be many con-
straints on their ability to do so. Therefore, whether child care and maternal
employment in the pre-school years are beneficial or harmful, and for which
children, are ultimately empirical questions. In the following sections, I review
the evidence on these questions. But first, it is important to establish some
ground rules for the analysis.

2. Ground rules for the analysis

The first ground rule is that one must be clear about what type of child care
one is analysing. There is a good deal of evidence about the benefits of early
childhood interventions. However, early childhood intervention and child care
are not synonymous. Early childhood intervention refers to programs such as
child care or home visiting that are designed to promote the development of
children from birth through the time they enter school and that typically are
targeted to children identified as high-risk for poor development. Child care,
in contrast, is not always designed primarily as an early childhood interven-
tion, and may be targeted to other groups (for instance, the children of em-
ployees or students). Child care is very heterogeneous, with provision ranging
from informal providers of care such as family members and babysitters to
more formal settings such as nursery schools, day-nurseries, and pre-schools.
Moreover, we do not know very much about the quality of child care being
o¤ered in most child care settings.4 Yet, there is evidence that quality of child
care matters for child outcomes.5 Thus, in reviewing any study of child care
and child outcomes, it is important to establish what type of child care was
provided and what, if anything, we know about the quality of that care. It
is also important to think about what the child care was meant to provide;
some programs, for instance, place more weight on cognitive development
than others.
The second ground rule is that one must be clear about when the child care
or maternal employment occurred. In the case of child care, there is a great
deal of evidence that child care begun in the first year of life has a di¤erent
e¤ect on later emotional adjustment than care begun thereafter (Haskins 1985;
Belsky and Eggebeen 1991; Baydar and Brooks-Gunn 1991; Smith 1994; Bates
et al. 1994; Ho¤erth 1999). The same may be true of cognitive development.
Child care begun in the first year of life has been found to have negative e¤ects
in several studies (Desai et al. 1989; Baydar and Brooks-Gunn 1991; Blau and
Grossberg 1992; Smith 1994; Han et al. (2001); Ruhm 2000; Waldfogel et al.
530 J. Waldfogel

2000); in contrast, care after the first year of life seems to have positive e¤ects
(Baydar and Brooks-Gunn 1991; Blau and Grossberg 1992; Brooks-Gunn et
al. 1992; Brooks-Gunn et al. 1993; Han et al. (2001); Waldfogel et al. 2000).6
The few studies that have been able to control for child care quality find that it
plays an important mediating role (Vandell et al. 1988; Field 1991; NICHD
Early Child Care Research Network 1997, 1999, and 2000; Burchinal et al.
1998), as does the type of care (Howes 1988 and 1990; Baydar and Brooks-
Gunn 1991; Field 1991; Smith 1994; Ho¤erth 1999). It may also matter
whether the care was full-time or part-time (Belsky and Eggebeen 1991; Har-
vey 1999; Han et al. (2001); Waldfogel et al. 2000).
The third caution is that one must be clear about which children experi-
enced the child care or maternal employment. Again using child care as an
example, the age at which a child enters non-maternal child care is obviously a
critical mediating factor, but so too are factors such as the child’s attributes,
family background, current living situation, and the quality of care provided
by the child’s own parents. These characteristics may influence both the type
of child care used and the child’s outcomes; thus, if child and family charac-
teristics are not properly controlled, one may erroneously attribute outcomes
as the result of child care when they are in fact the result of other factors.
Further complicating the analysis is the fact that child care and family char-
acteristics may have an interactive e¤ect. For instance, the NICHD study of
early child care in the U.S. found that infants whose parents had more sen-
sitive childrearing styles fared better than other children in early child care
(NICHD Early Child Care Research Network 1997), while several studies
have found that children from families that are economically disadvantaged
gain more from child care in terms of their cognitive development than do
other children (see, for example, Desai et al. 1989; Vandell and Ramanan
1992; Brooks-Gunn et al. 1992; Brooks-Gunn et al. 1993; Caughy et al. 1994).
The fourth point is that one must be clear about what outcomes one cares
about. To a large extent, the outcomes one tracks will depend on the type of
child care being considered, the time at which it was delivered, and the type of
children who received it, but it is important to remain open to unanticipated
outcomes as well. Thus, in tracking the e¤ects of early child care, it is natural
to focus on issues of separation and attachment, but it would be useful to look
at later social and cognitive outcomes as well. And, in assessing cognitively-
oriented programs for older pre-schoolers, it makes sense to look at school
outcomes, but it is also important not to lose sight of other outcomes that
may be a¤ected. Implicit in this discussion is the notion that it makes sense
to look at long-term as well as short-term outcomes, and at potential benefits
for society as a whole in addition to those that may accrue to the child and his
or her family.
With these ground rules in mind, let us now turn to the evidence on the
potential benefits, and the potential ill e¤ects, of child care and maternal em-
ployment in the pre-school years. We examine first the literature on the po-
tential benefits of child care programs that are designed as early interventions.

3. Potential benefits of early intervention child care programs

We now know a good deal about the potential benefits of early intervention
programs, including programs that provide child care. Much of the evidence
comes from research conducted in the United States, so I begin there.
Child care, women’s employment, and child outcomes 531

There have been several excellent reviews of the U.S. research on early
childhood interventions and outcomes. The most recent, and the most useful
for the purposes of this paper, is the RAND study which rigorously assessed
nine early intervention programs, including eight child care programs (Karoly
et al. 1998).7 In order to be included in the RAND review, studies had to meet
high scientific standards; in particular, they had to have used random assign-
ment or other techniques to control for pre-existing di¤erences between treat-
ment and controls and they had to follow the treatment and control groups
over time so that they could assess long-term as well as short-term outcomes.
The results of the RAND review, summarized in Table 1, show that well-
designed early intervention programs, including child care programs, can make
a positive di¤erence in the lives of children. The results also show that the ef-
fects of programs vary by what specific type of program was o¤ered. All eight
child care programs included in the RAND review were cognitively oriented
and all of these programs were successful at raising children’s cognitive test
scores or school achievement as measured by higher IQ scores, higher school
achievement test scores, less time in special education, better grades, less grade
repetition, or higher rates of graduation from high school (the one exception,
the Elmira PEIP, was a parental support program that used a home visiting
model designed to reduce abuse and neglect). But the gains of these programs
were not limited to cognitive outcomes. The High/Scope Perry Pre-School
Project, for instance, led to higher employment, earnings, and income; it also
led to lower rates of crime and delinquency, as did two other programs (the
Syracuse FDRP and the Chicago CPC programs). Interestingly, although
most programs were child-focused, many were successful at changing parents’
behaviours in positive ways: the Elmira PEIP home visiting program reduced
abuse and neglect and parental welfare use; the Houston PCDC and the IHDP
home visiting and day care programs improved mother-child interaction and
the HOME score (an index of how well the home environment promotes
child development); the Syracuse FDRP home visiting and day care program
and the Carolina Abecedarian program raised mothers’ level of education;
the Carolina Abecedarian and IHDP programs raised maternal employment;
and the Chicago CPC day care and follow-through program raised parents’
involvement in their child’s school.8 Some of these e¤ects on parents were
intended but most were not.
Program outcomes varied by when services were delivered. In general,
programs that intervened earlier and that were more intensive (such as
Carolina Abecedarian and IHDP) had stronger e¤ects than those that inter-
vened later and less intensively, and programs (such as Carolina Abecedarian
and the Chicago Child-Parent Centers) that included a follow-through com-
ponent were more successful at sustaining gains than those that didn’t (see
also Reynolds 1998 on this point).
Consistent with prior research, some programs were more beneficial for
higher-risk children. For instance, the IHDP program produced the greatest
IQ gains for the children whose parents had the least education themselves.
Compared to control children whose parents had the same level of education,
children whose mothers had less than a high-school education gained nearly
20 IQ points by age 3, children whose mothers were high-school graduates or
had some college education gained about 10 IQ points, while children whose
mothers were college graduates gained no IQ points (Ramey and Ramey
1998b).
532
Table 1. The e¤ects of early childhood interventions: Selected U.S. studies
Program
Prenatal/Early Infancy Project
(PEIP)
Elmira, New York, 1978–1982
N ¼ 304, Random Assign-
ment (RA)
First births to young, single,
or low SES mothers, served
ages 0–2.
Home visiting.
Early Training Project (ETP)
Murfreesboro, TN, 1962–1965
N ¼ 65, RA
Low SES children, ages 4–6.
Summer part-day (PD) pre-
school & home visiting.
High/Scope Perry Pre-School
Ypsilanti, Michigan, 1962–
1967
N ¼ 123, RA
Low SES & low IQ, ages 3–5.
School-year PD pre-school &
home visiting.
Houston Parent-Child
Development Center
(PCDC)
Houston, Texas, 1970–1980
N ¼ 291, RA
Low SES, ages 1–3.
PD day care & home visiting.
Syracuse Family Development
Research Program (FDRP)
Syracuse, New York, 1969–
1975
N ¼ 216, control group but
not RA
Low SES, Ages 0–5.
PD (for infants) & full-day
(FD) family day care &
home visiting.
J. Waldfogel
Statistically significant No statistically significant
di¤erence between di¤erence between
treatments and controls treatments and controls
Emergency room visits at IQ at age 3 and 4.
ages 2–4. HOME score at age 4.
Crime/delinquency by age Mom’s education by age 4.
15.* Mom’s employment by age 15.
Reports of abuse/neglect by
age 15.
Mom’s welfare use by age 15.
IQ at age 6. IQ at age 7 and 17.
Achievement at age 7. Achievement at age 10 and 17.
Special education by age 18. Grade repetition by age 18.
High school (HS) completion HS completion by age 18.
after pregnancy by age 18. Teen pregnancy by age 18.
IQ at age 5 and 7. IQ at age 8 and 14.
Achievement at age 9 and 14. Teen pregnancy by age 19.
Employment at age 19. Grade repetition by age 27.
Special education by age 19 Post-HS education by age 27.
and 27. Employment at age 27.
HS completion by age 27.
Crime/delinquency by age 27.
Income at age 27.
Welfare participation at age
27.
Earnings at age 27.
IQ at age 2. IQ at age 3.
Mother-child interaction at Special education at ages 8–11.
age 3. Grade repetition at ages 8–11.
HOME score at age 3. Grades at ages 8–11.
Behavior at ages 4–7.
Achievement at ages 8–11.
Bilingual education at ages
8–11.
IQ at age 3. IQ at age 6.
Behavior at age 3. Behavior at age 6.
Mom completed HS by age 5. Special education by age 15.
Crime/delinquency at age 15. Grade repetition by age 15.
Grades at age 15.**
School attendance at age
15.**
Teacher ratings at age 15.**
Referred by probation by age
15.**
Child care, women’s employment, and child outcomes 533

Table 1. (continued)

Program Statistically significant No statistically significant

di¤erence between di¤erence between
treatments and controls treatments and controls

Carolina Abecedarian IQ at age 5 HOME score at age 5.

1 site in NC, 1972–1985
N ¼ 117, RA
High-risk families, ages 0–8.
FD year-round center-based
educational day care for
pre-schoolers, followed by
parent program for school-
age kids.
Project CARE (Carolina
Approach to Responsive
Education)
1 site in NC, 1978–1984
N ¼ 65, RA
High-risk families, ages 0–5.
Home visiting & FD year-
round center-based
educational day care, or
home visiting only.
Infant Health and
Development Project
(IHDP)
8 sites, 1985–1988
N ¼ 985, RA
Premature & low birth weight
(LBW) infants, ages 0–3.
Home visiting for infants
followed by FD year-round
center-based educational
day care.
Chicago Child-Parent Center
(CPC)
Chicago, Illinois, 1967-present
N ¼ 1539, statistical controls
Low SES, ages 3–9.
PD pre-school followed by
FD kindergarten followed
by extra support in
classroom and after school
in primary grades.
Mom’s education by age 5. IQ at age 15.
Mom’s employment by age 5.
IQ at age 8.
Achievement at age 8.
IQ at age 12.
Achievement at age 15.
Special education by age 15.
Grade repetition by age 15.
IQ at age 1. Childrearing attitudes at age 3.
IQ at age 3. HOME score at age 5.
IQ at age 5.
IQ at age 3. Mom’s education by age 3.
Behavior at age 3. Time on welfare by age 3.
Mother-child interaction at Subsequent pregnancy by age 3.
age 3. Behavior at age 8.
HOME score at age 3. Grade repetition by age 8.
Mom’s employment at age 3. Special education by age 8.
Behavior at age 5.***
IQ at age 5.***
IQ at age 8.***
Math achievement at age
8.***
Achievement at age 9. Behavior at age 9.
Parents involved in school at Crime/delinquency by age 16.
age 9.
Achievement at age 14.
Grade repetition by age 14.
Special education by age 14.
Crime/delinquency by age 14.

Source: Karoly et al. 1998; Tables 2.1, 2.2, and 2.3.

* Di¤erence statistically significant for the high-risk group only.
** Di¤erence statistically significant for girls only.
*** Di¤erence statistically significant for heavier LBW children only.

The Rand study did not include Head Start because no Head Start
evaluation met the Rand criteria for scientific rigor. However, Head Start is
an important example: it is the single largest American child care program
and probably the best known. Early studies of Head Start concluded that
the program had positive e¤ects on children’s cognitive abilities and school
534 J. Waldfogel

achievement but these e¤ects seemed to ‘‘fade out’’ over time (see, for instance,
McKey et al. 1985). However, the most recent evidence on Head Start reveals
a more nuanced story (Lee et al. 1990; Currie and Thomas 1995, 1996a, and
1996b). Children who attended Head Start have higher test scores at the end
of the program than siblings who stayed at home or attended some other type
of pre-school. Head Start children are also more likely to be immunized than
siblings who stayed home. While the test score e¤ects for African-American
children fade out fairly rapidly, perhaps because they go on to attend poor
schools, the e¤ects for white and Hispanic children are longer-lasting. White
and Hispanic children who attended Head Start have higher test scores at
age 10 than comparable children who did not attend Head Start. White Head
Start children are also less likely to have repeated a grade by age 10 than
comparable white children who did not attend Head Start.
Head Start continues to enjoy broad public and bipartisan support in the
United States, and the program is now being expanded in two directions. First,
Early Head Start is now delivering Head Start services to children under the age
of three, reflecting the new emphasis on interventions in the first three years of
life (and also reflecting the fact that older pre-school age children are increas-
ingly likely to be served by the public schools or other pre-schools).9 Second,
Head Start Follow-Through programs are now following Head Start children
into the school years, to see whether Head Start gains can be better maintained
if follow-through services are provided (as they were in the Carolina Abece-
darian program and the Chicago Child and Parent Centers).
As noted above, much of the evidence on early intervention programs
comes from the U.S., but there is some evidence from other countries that
should be mentioned. In particular, it is important to mention three countries
that have mounted large-scale early intervention initiatives – France, Ireland,
and Britain.
France is well-known for its ecoles maternelles, publicly-run and -funded
pre-schools which enroll nearly 100% of pre-school children age three and
older, and an increasing share of children age two (Boocock 1995). Although
these pre-schools are renowned as a source of universal and high quality public
child care, they were also intended to serve as a form of compensatory educa-
tion, with the aim of reducing the gap in school readiness between the better-
o¤ and the less well-o¤. Longitudinal research in France confirms that they
have met this goal: children who attended these pre-schools for more years do
better in school (McMahan 1992; see also studies reviewed in Rayna and
Plaisance 1998).
Ireland embarked on its ambitious national early intervention program,
called Early Start, in 1994 (Hayes 1998). This program di¤ers from the uni-
versal French program in that it is targeted to children living in disadvantaged
areas. It provides a part-day educational enrichment program to 3- and 4-year
olds, drawing upon lessons from research from other countries and also upon
Ireland’s own successful Rutland Street pre-school project. The Rutland Street
project, which served disadvantaged children in Dublin starting in 1969, was
found to produce short-term gains in children’s test scores and long-term gains
in the rates at which children stay in school and take exams for higher edu-
cation (Kellaghan and Greaney 1993).
Britain has a long tradition of early intervention e¤orts which have
proven e¤ective in improving children’s and families’ outcomes in a number
of domains (for reviews, see Sylva and Wiltshire 1993; Ball (ed.) 1994; Oliver
Child care, women’s employment, and child outcomes 535

and Barker 1998; and Zoritch and Roberts (forthcoming). Drawing on its own
research base, as well as lessons from U.S. research, Britain is now mounting a
major early intervention e¤ort, the Sure Start program. Like Early Start, Sure
Start is targeted to children living in disadvantaged areas, but it di¤ers in that
it is more comprehensive, o¤ering a wide range of supportive services to chil-
dren and families, and also serves a broader age band of children, beginning
prenatally and following children to at least age four (Glass 1999).10
The programs reviewed above were for the most part designed as early
interventions. But, as noted earlier, child care and early intervention program
are not synonymous. So, while the positive evidence on early intervention child
care programs is reassuring, it does not address the question of the potential
benefits or ill e¤ects of having a working mother and participating in other
forms of non-maternal child care during the pre-school years. We turn to these
topics next.

4. E¤ects of maternal employment and child care in the pre-school years

As discussed above, there is a large body of research on the e¤ects of maternal
employment and child care in the pre-school years on child outcomes.11 Much
of this research has focused on the question of whether early maternal em-
ployment and child care, in particular employment and child care begun in the
first year of the child’s life, have adverse outcomes for children. The focus
of the early research on this topic was on socioemotional rather than cogni-
tive outcomes, with a particularly vigorous debate about attachment. Several
studies found di¤erences in attachment between children who had been in
child care in the first year of life and those who had not, but experts disagreed
about how to interpret these results. If children who had been in child care
as infants engaged di¤erently with their mothers, this might be a symptom of
attachment problems (Belsky 1988) or it might be a mature, adaptive response
to the child care experience (Clarke-Stewart 1988). Nor was it clear how such
attachment di¤erences might a¤ect later outcomes.
This line of research, and the associated debate, dominated the child
care research agenda in the United States for many years, as study after study
tended to find negative e¤ects of child care and maternal employment in the
first year of life for at least some groups, as compared with generally positive
e¤ects of child care and maternal employment after the first year of life. There
has been a very active debate on these topics in Britain as well (McGurk et al.
1993; Morgan 1996). Studies in Britain produced mixed results about socio-
emotional development; for instance, Osborn and Milbank (1987) using data
on children born in 1970 found negative e¤ects of pre-school age child care
on socioemotional development but Melhuish and Moss (1991) using data
on children born in 1983 did not. The results for cognitive development
have been more consistently positive, with both Osborn and Milbank (1987)
and Melhuish and Moss (1991) reporting positive e¤ects of pre-school child
care.
This research, however, left two important questions largely unanswered.
The first is whether any negative e¤ects of early maternal employment and
child care are lasting. The second concerns the extent to which the e¤ects of
child care and maternal employment in the first years of life vary according to
characteristics of the child, the family, and the caregiver. Thus, more recently,
researchers have turned their attention to these two sets of questions.
536 J. Waldfogel

4.1. Are the e¤ects of early maternal employment lasting?

Eight studies have been published to date examining the impact of early
maternal employment on child outcomes in the U.S. using data from the
National Longitudinal Survey of Youth (NLSY). However, with one excep-
tion, all these studies have analyzed children at young ages or at one point in
time.
Desai et al. (1989) analyzed 503 children who were 4 years old in 1986,
combining whites, blacks, and Hispanics, but conducting separate analyses for
boys and girls; they found that high income has a significant positive e¤ect on
cognitive development as measured by scores on the Peabody Picture Vocab-
ulary Test (PPVT-R), but being in a high-income family where the mother
worked in the first year had a negative e¤ect on PPVT-R scores, especially
for boys. Baydar and Brooks-Gunn (1991) analyzed 572 non-Hispanic white
children aged 3 and 4 in 1986 and found negative e¤ects of early maternal
employment on behavioral adjustment (as measured by the Behavior Problem
Index, BPI) and on cognitive development (as measured by the PPVT-R).
Belsky and Eggebeen (1991) studied 565 white and black children aged 4–6
in 1986 and found that children whose mothers were employed full-time in the
first or second year of life scored worse on a composite measure of behavioral
adjustment than did children whose mothers were not employed during their
first three years. Blau and Grossberg (1992) studied 876 children who were 3–
4 years old in 1986 and found a significant negative e¤ect on PPVT-R scores
of the mother working more weeks in the first year of life; however, this e¤ect
was compensated if the mother worked more weeks in the second and third
years of life. Vandell and Ramanan (1992) studied 189 low-income children
who were in second grade in 1986 and found that maternal employment had a
positive e¤ect on PIAT-Math scores but a negative e¤ect on PPVT and PIAT-
Reading scores. Parcel and Menaghan (1994) studied children aged 4–6 in
1986 (786 with PPVT-R scores and 526 with BPI scores) and found minimally
negative e¤ects of early maternal employment on child cognitive outcomes but
no e¤ect on children’s behavioral problems. Greenstein (1995) analyzed 2,040
children age 4–6 in 1986, 1988, or 1990, conducting separate analyses by race/
ethnicity and by gender; he hypothesized that the adverse e¤ects of early
maternal employment might exist only for the most advantaged children, but
his results did not confirm this.
Harvey’s (1999) study is exceptional in that it assesses some outcomes as
late as age 12 and at several points in time. However, Harvey’s study is not
longitudinal in design; she does not follow one sample of children over time.
Thus, when she finds that e¤ects that are present at one age are not present at
another age, one does not know whether the e¤ects have attenuated or whether
they are simply present for one group of children but not another. Another
limitation of Harvey’s study is that she analyzes white, African-American, and
Hispanic children together. If there are important di¤erences in the e¤ects
of early maternal employment across racial and ethnic groups, as some prior
research has found, then analyzing children separately is likely to yield more
accurate estimates. Further complicating the interpretation of Harvey’s results
is the fact that she does not control for family income. As noted earlier,
maternal employment may have positive e¤ects on child outcomes by raising
family income. Thus, when Harvey finds essentially zero e¤ects of maternal
employment on later outcomes, it is possible that this result reflects the com-
Child care, women’s employment, and child outcomes 537

bination of positive e¤ects due to the increase in income and negative e¤ects
due to the absence of the mother.
Two recent longitudinal studies tackle some of these questions. The first
(Han et al. 2001) analyzed the e¤ects of early maternal employment for a
small cohort of 244 white and 218 African-American children from the NLSY
followed longitudinally from the time they were born in 1982 or 1983 to age
3–4 in 1986, age 5–6 in 1988, and age 7–8 in 1990. Analyzing the white and
African-American children separately and controlling for family income, the
authors found that maternal employment in the first year of a child’s life had
small but statistically significant negative e¤ects on cognitive outcomes for
white children and that these e¤ects persisted to age 7 or 8. However, they
found no e¤ects of first-year maternal employment on cognitive outcomes for
black children.
The second longitudinal study (Waldfogel et al. 2000) used a larger sample
of children from the NLSY – 903 whites, 582 African-Americans, and 387
Hispanics born between 1982 and 1989, followed from birth to age 7 or 8 –
and again found small but persistent negative e¤ects of first-year maternal
employment on later cognitive outcomes for white children, but not African-
American children (the results for Hispanic children were mainly inconclusive).
The authors tested several hypotheses to explain these negative e¤ects of first-
year maternal employment. In support of the hypothesis that maternal sepa-
ration causes poorer cognitive outcomes, they found that the adverse e¤ects of
first-year employment were larger for children whose mothers worked longer
hours in the first year and for children whose mothers began work before
the fourth quarter of the first year. At the same time, however, the fact that
these adverse e¤ects were found only for white children, and not for African-
American children, cast doubt on a pure maternal separation explanation. Also
challenging this explanation was the fact that the authors found no evidence
of lasting impacts of first-year maternal employment on behavioral problems
for whites or other children, even though socioemotional functioning presum-
ably would be most impacted by maternal separation. The authors also found
some support for a hypothesis emphasizing the role of fathers, in that the ad-
verse e¤ects of first-year employment were mainly concentrated in families in
which a mother worked and a father was present and were not found in single-
mother families, and some support for a breast-feeding hypothesis, in that the
e¤ects of first-year employment were strongest in families where the mother
was employed and did not breast-feed. The authors also tested for whether the
e¤ects of first-year maternal employment varied by the mothers’ own cogni-
tive ability level, or by the families’ income level, and found that the adverse
e¤ects were concentrated among families where mothers had moderate levels
of ability, and moderate levels of income. Because previous studies have found
that it is often moderate income children who are in the lowest quality child
care, since higher-income mothers are better able to obtain and purchase
high-quality child care, while lower-income mothers are often eligible for high-
quality subsidized child care (Phillips et al. 1994), these results may suggest
that the quality of child care in which a child is placed moderates the e¤ects
of maternal employment. However, Waldfogel et al. (2000) in common with
other NLSY studies, were unable to test this conjecture because the NLSY
contains no data on child care quality.
A further limitation of these NLSY studies is that they have not controlled
for potential biases that might arise due to the endogeneity of maternal em-
538 J. Waldfogel

ployment or child care choices. The endogeneity problem is that there may
be characteristics of the child, mother, or family that are correlated with the
outcomes researchers are concerned about but that are not recorded in the
dataset. To the extent that these unobserved characteristics influence maternal
employment or child care decisions, then analyses that do not control for them
may erroneously attribute their e¤ect to maternal employment or child care
(this is an instance of the familiar problem of ‘‘omitted variables bias’’). For
instance, if mothers who are less-skilled at boosting their children’s cog-
nitive skills are also more likely to return to work early, then studies might
erroneously attribute the e¤ect of those lower skills to early employment, and
thus the estimated e¤ects of early maternal employment would be downwardly
biased. Or, the e¤ects might go in the other direction. If, for instance, children
who are more alert and responsive have mothers who decide that it would be
all right for them to go to child care early, then estimates of the e¤ects of early
child care on cognitive outcomes could be upwardly biased.
Unless researchers have an experiment that randomly assigns families to
the early employment and child care group or to a control group, or an ‘‘in-
strument’’ (something beyond the mother’s control that demonstrably influ-
ences her choices), it is not possible to determine which way the bias goes or to
correct for it. Therefore, the best one can do is to control for as many relevant
characteristics of the child, mother, and family as possible, in the hopes of re-
ducing the omitted variables problem, and the potential for bias, to a minimum.
A recent example of such an approach is a study by Ruhm (2000) who uses an
expanded set of covariates from the NLSY in an attempt to control for more of
the characteristics that might otherwise bias the estimated e¤ects of maternal
employment on child outcomes. His results suggest that when one does con-
trol for a richer set of covariates, the impact of early maternal employment on
cognitive outcomes becomes more negative. However, because he uses data
from the NLSY, there are still many important characteristics of the family
and child care environment that he is not able to control for.
As noted earlier, there has been an active debate on the e¤ects of early
maternal employment in Britain, and two sets of longitudinal studies have
recently been carried out there. In a program of research sponsored by the
Joseph Rowntree Foundation, John Ermisch and Marco Francesconi have
investigated the impacts of parental employment on outcomes for young
adults who were born in the 1970s and were included in the British Houshold
Panel Study in the early to mid-1990s. Although earlier studies (Francesconi
and Ermisch 1998a, b) had found that the e¤ects of maternal employment
for this sample were generally positive, in this more recent research, Ermisch
and Francesconi (2000) disaggregated maternal employment by the age of the
child and the intensity of hours worked and found that children whose mothers
worked more extensively during their pre-school years had somewhat lower
educational attainment as young adults and that these e¤ects held up (in fact,
become even stronger) when they controlled for unobserved heterogeneity
among mothers by estimating family fixed e¤ects models. The second pro-
gram of longitudinal research was carried out by Heather Joshi and Georgia
Verrropoulou, using data from two large national birth cohort studies (the
National Child Development Study of children born in 1958 and the British
Cohort Study of children born in 1970), with funding from the Smith Institute.
Joshi and Verropoulou (2000) first examined outcomes for children born in
the 1970s and 1980s to mothers from the 1958 birth cohort and found that
Child care, women’s employment, and child outcomes 539

these children tended to have slightly poorer outcomes when assessed at ages
5 to 17 in 1991 if their mothers worked in the first year of life, although only
the e¤ect on children’s reading performance was statistically significant. They
next examined children born in 1970 and found, consistent with the results
of Ermisch and Francesconi (2000) from a di¤erent dataset but from roughly
the same period, that children whose mothers worked more hours during the
pre-school years had somewhat lower educational attainment as young
adults.12
Longitudinal studies of the impact of early maternal employment and early
child care in other countries are less common. An important exception is the
work on the long-run e¤ects of early child care in Sweden, which finds that
children who had been in day care before the age of two had better socio-
emotional and cognitive outcomes at ages 8 and 13 than children who had
not attended out of home care, with the best outcomes for children who began
day care before age one (Andersson 1989, 1992; see also Broberg et al. 1989,
and Cochran and Gunnarson 1985). As noted above, there has also been some
longitudinal research on early child care in France, where there has been an
active debate about the impacts of child care for two year olds, with the results
of several studies now indicating that the e¤ects on cognitive achievement
are generally positive (see Boocock 1995 and Rayna and Plaisance 1998 for
reviews). However, to a large extent, research in the Nordic and continental
European countries has been less concerned with studying whether children
should be in child care, given the high share of children who are already in
such care, than with understanding how to make that experience as beneficial
as possible in terms of promoting positive child outcomes, taking into account
child and family characteristics as well as the characteristics of the child care
setting. We turn to this set of questions next.

4.2. How do the e¤ects of early child care vary by child, family,
and child care characteristics?

The NICHD Study of Early Child Care has brought together many of the
leading developmental psychologists in the U.S., including prominent rep-
resentatives from both sides of the attachment debate, to conduct a unique
national longitudinal study of the e¤ects of early child care on child outcomes
for a diverse sample of 1,374 children from 10 sites across the country. Al-
though this study is limited in that it is a naturalistic rather than experimental
study and thus does not allow researchers to control for the potential endo-
geneity of maternal employment and child care choices, it is an improvement
on the NLSY in that it contains richer data on child, family, and child care
characteristics. These data, while by no means eliminating the endogeneity
problem, may reduce the extent of the biases associated with omitted variables.
Results from this study, which is still ongoing, are summarized in Table
2.13 This study has provided new evidence on the e¤ects of early child care,
finding that child care per se neither helps nor harms attachment, but also that
longer hours in care in the first six months of life may pose risks in terms of
mother-child relationships and child behavior problems. The NICHD study
has also provided new evidence that the quality of care matters, finding signif-
icant e¤ects of higher quality care on mother-child relationships, on children’s
cognitive and language development, and on their school readiness.
540 J. Waldfogel

Table 2. Results to age 3 from the NICHD study of early child care

. Childcare per se neither helps nor harms attachment.

For children whose mothers are sensitive, childcare has no e¤ect on attachment. For children
whose mothers are not sensitive, high quality care leads to more secure attachment, while poor
quality care, more than 10 hours per week of care, or more than 1 care arrangement by age 15
months leads to less secure attachment.
. Longer hours of care in the first six months may pose some risk.
Longer hours are associated with more behavior problems at age 2 and lower maternal
sensitivity and less positive interactions at age 3. But, child and family characteristics are more
important.
. Quality of care has an e¤ect on mother-child relationships.
Higher quality care predicts greater maternal involvement and sensitivity at 15 and 36 months
and more positive interactions at 36 months. Low-income mothers using high-quality care
have more positive interactions with their children at age 6 months than those who do not use
care or who use lower-quality care.
. The quality, type, and consistency of childcare in the first three years of life also have e¤ects
on children’s socioemotional outcomes.
Time spent in quality childcare, and time spent in group care, were associated with children
displaying greater social competence and cooperation and fewer problem behaviors at ages 2
and 3. At age 2, children who had been in many childcare arrangements showed more problem
behaviors.
. The quality of childcare in the first three years of life a¤ects children’s cognitive and language
development and their school readiness.
The higher the quality of care – in terms of language stimulation and child-caregiver
interactions – the higher the child’s language skills at 15, 24, and 36 months. Higher quality
care also is associated with cognitive development at age 2 and school readiness at age 3.
Children in day care centers that meet quality standards across all four domains assessed –
child-sta¤ ratio, group size, teacher training, and teacher education – have better language
comprehension and school readiness, and fewer behavior problems, than children whose
centers fail to meet the standards in all four domains.
. The overall e¤ect of childcare depends on the quality of that care.
High quality childcare o¤ers an advantage to children and low quality care a disadvantage to
children in terms of their cognitive and language development compared to care from the
average mother.

Source: NICHD Early Child Care Research Network, 1996, 1997, 1998, 1999, 2000.

The NICHD results on school readiness are reinforced by the newly released
results from the Cost, Quality, and Outcomes follow-up study (see Table 3)
which followed 418 children from pre-school to second grade (i.e., from age
4 to 8). Although this study too does not control for the potential endogeneity
of child care choices, its results provide some evidence that those who at-
tended higher quality pre-schools had better outcomes in second grade than
those who attended lower-quality pre-schools, with particularly strong results
for the children of less-educated mothers.
The results from this latest wave of research suggest that one can not make
sweeping conclusions about whether early child care harms, or helps, children;
rather, the e¤ects of early child care on a child’s attachment, child-mother
interactions, and cognitive and behavioral outcomes depend critically on the
characteristics of that care (including the quality of the care, its continuity,
and the number of hours that the child is in care) and the characteristics of
the child and family. Thus, increasingly, interest is shifting from the question
of whether early child care or maternal employment harms children to the
question of what types of early child care can be most helpful for what types
of children.
Child care, women’s employment, and child outcomes 541

Table 3. Results to age 8 from the cost, quality, and outcomes follow-up: ‘‘The children of the
cost, quality, and outcomes study go to school’’

. Children who attended higher quality childcare centers performed better on measures of
cognitive and social skills in childcare and in kindergarten (after they made the transition to
school).
. The quality of childcare classroom practices had a positive e¤ect on children’s cognitive
development (as measured by language and math skills), while the closeness of the childcare
teacher’s relationship with the child had a positive e¤ect on children’s social development
(as measured by thinking/attention skills, sociability, lack of behavior problems, and peer
relations).
. The e¤ects of childcare quality on math ability, thinking/attention skills, and problem
behaviors persisted through second grade. Childcare quality also had an e¤ect on peer
relations in second grade, with children having better relations with peers if they had attended
childcare classrooms with fewer problem behaviors, closer teacher-child relationships, and
more opportunities for children to play together.
. For two outcomes – math skills and problem behaviors – the e¤ects of high-quality childcare
were greater for children whose mothers had less education than for children whose mothers
had more education.

Source: Peisner-Feinberg et al. 1999, ‘‘Executive Summary of Cost, Quality, and Outcomes
Study’’.

This approach can be seen in a new British study, the E¤ective Provision
of Pre-School Education (EPPE) Project, being conducted by Kathy Sylva
and colleagues with funding from the British Department for Education and
Employment (see Sylva et al. 1999, 2000). This longitudinal study is follow-
ing 3,000 children, from 141 pre-school settings, from age three to age seven,
as they move from pre-school into primary school and is gathering data on
each child care setting attended by the children, with the aim of investigating
the e¤ects of child care characteristics, as well as individual and family char-
acteristics, on children’s developmental progress. Although it is the first large-
scale study to investigate these topics in Britain, the EPPE project builds on
several earlier British studies of the e¤ects of various types of preschool pro-
vision. These studies have found considerable variation in the e¤ect of pre-
school depending on the type of provision (Melhuish et al. 1990; Melhuish
1993; Sylva and Wiltshire 1993). The e¤ects of child care also seem to vary
by time period (see, for instance, Feinstein et al. (1998) who found di¤erent
e¤ects of di¤erent types of pre-school care for children born in 1958 and 1970).
Interestingly, the type of research being carried out in the newer U.S. and
British studies is quite consistent with the type of research that has dominated
the early childhood research agenda in the Nordic and continental European
countries (see, for instance, studies of the impact of child care quality in
Sweden, such as Hwang 1990 and Broberg et al. 1997, or studies of the e¤ects
of di¤erent types of care on di¤erent types of outcomes for children in France,
reviewed in Rayna and Plaisance, 1998). These countries, as we have seen,
have been less occupied with questions about whether mothers should work
or whether children should be in child care and more interested in questions
as to the right type of child care for particular types of children. This focus
probably reflects the fact that for some time now, a high proportion of Nordic
and continental European preschoolers have attended some form of child
care. Thus, the questions of interest tend to be concerned with how to make
that experience as positive as possible, by improving the curriculum, the match
between child and care setting, and the links between care settings and parents
542 J. Waldfogel

(see, for instance, Brostrom and Vilien 1998, on the research agenda in Den-
mark, or Rayna and Plaisance 1998, on the research agenda in France).

5. What we don’t know about child care, maternal employment,

and child outcomes

We still have much to learn about the e¤ects of child care and women’s
employment on child outcomes. In this concluding section, I want to highlight
knowledge gaps in three areas.
First, many questions remain to be answered about the e¤ectiveness of
early childhood child care interventions. The RAND review conducted by
Karoly and colleagues (1998) highlighted five important limitations in terms
of our current knowledge about early childhood interventions, including child
care interventions: 1) we don’t know enough about optimal program designs
(e.g., should programs focus on parents, children or both, should programs
begin in infancy or later in the pre-school years, should programs o¤er spe-
cialized services to a small number of children or generic services to a larger
number of children); 2) we don’t know enough about how to target inter-
ventions to those who would benefit most (in some cases, such as programs
to boost school achievement, we do know that children whose mothers have
lower school achievement themselves will benefit most and therefore we can
target e¤ectively, but in other cases, such as programs to improve social func-
tioning in areas such as abuse and neglect, we may not know which families
to target or the merits of targetting to high-risk families may be o¤set by the
damage caused by stigmatizing participants); 3) we don’t know whether the
types of programs that have been favorably evaluated can be replicated on
a larger scale (this is the ‘‘scale-up’’ problem); 4) we don’t know what the full
range of program benefits might be, since programs have typically been eval-
uated in terms of only a subset of benefits (for instance, cognitive gains but
not other gains for the child, or benefits for the child but not the parents); and
5) we don’t know whether the benefits that have been found in earlier studies
will be the same in di¤erent policy environments (for instance, how well will
results hold up in the U.S. as welfare reforms are implemented, and to what
extent are results from the U.S. generalizable to other countries). The evalua-
tions of the Early Head Start initiative in the U.S., and the Sure Start initiative
in Britain, should provide evidence on some of these points (in particular, on
the scale-up question and on the di¤ering policy environments question) but
on each of these, more evidence and more analysis is needed.
Second, there is still much to be learned about the long-run e¤ects of early
child care and maternal employment on child outcomes, and on why and how
these e¤ects vary across children. Although many studies of the e¤ects of early
child care and maternal employment have been conducted using the NLSY,
only two have been longitudinal in design. Further longitudinal studies with
the NLSY, and with other datasets, are needed if we are to understand the
long-run e¤ects of early child care and maternal employment. There is also
ample scope for further analyses of the rich NICHD data on early child care
if we are to understand how and why the e¤ects of early child care and ma-
ternal employment vary by child, family, caregiver, and so on. As discussed
above, an important limitation of both the NLSY, and NICHD studies, is
Child care, women’s employment, and child outcomes 543

that neither set of studies has been able to tackle the endogeneity problem.
Thus, drawing causal interpretations as to the impact of employment or child
care choices on child outcomes remains problematic.
Third, we have probably not learned as much as we might about child
care, women’s employment, and child outcomes in non-U.S. settings. Most
of the evidence I have cited comes from the United States, but this evidence
may or may not be applicable to other settings. To take Britain as an example,
although as we have seen there have been several recent reviews of the e¤ects
of early childhood interventions and child care on child outcomes (see for in-
stance Sylva and Wiltshire 1993; Ball (ed.) 1994; Oliver et al. 1998; and Zoritch
and Roberts (forthcoming)), and several studies of the e¤ects of women’s em-
ployment on child outcomes (see for instance Ermisch and Francesconi 1999
and Joshi and Verropoulou 2000), the research base on child care in the pre-
school years is still relatively small. Yet, in Britain as in many other indus-
trialized countries, employment among mothers of infants is increasing more
rapidly than for any other group, and an increasing share of infants is in non-
maternal care.14 As Lynch (2000) has noted in the U.S. context, these trends
present both a challenge and an opportunity, and the outcomes for children
will depend to a large extent on the type and quality of the care they receive.
Yet we know very little currently about what forms of child care British
mothers of young children are using for their infants and toddlers, and the
quality of that care. We do not know which British children begin care early,
how young they are when they begin care, and how many hours a week they
are in care. And we do not know enough about the e¤ects of child care and
other early childhood interventions as delivered in Britain on outcomes for
children. Although we can learn a great deal from carefully conducted re-
search in other countries, we need to be careful to compare like to like. We
noted earlier that child care is very heterogeneous, and of course there is even
more variation across countries than there is within them. Moreover, the e¤ects
of child care may also be sensitive to the broader policy context. For instance,
we have seen in recent research that the long-run e¤ects of pre-school inter-
vention may depend on how supportive the child’s later school is and on
whether follow-through programming is provided (see for instance Reynolds
1998). Thus, longitudinal research on British children, receiving British early
childhood interventions and then entering British schools, is essential if one
wants to know which early childhood interventions would be most e¤ective
and whether follow-through programming will be necessary to ensure that
e¤ects do not fade out over time. The new EPPE study, discussed earlier,
should be very useful in this regard.
In conclusion, then, we now have a good deal of evidence that early child-
hood interventions, including child care interventions, can make a di¤erence
in improving outcomes for children. We also have a fair amount of evidence
on the e¤ects of early maternal employment and child care on outcomes for
children, and we are learning more about whether these e¤ects are lasting, and
on how they vary by child, family, and child care provider. However, there are
at least three important knowledge gaps that future work by economists could
fruitfully seek to address. We have more to learn about the benefits of early
intervention child care programs, about the impacts of early child care and
maternal employment on child outcomes, and about the relationships between
child care, women’s employment, and child outcomes in countries other than
the United States.
544 J. Waldfogel

Endnotes

1 On the importance of the early years, see the Carnegie Task Force on Meeting the Needs
of Young Children (1994) and Shore (1997). But see also Bruer (1999) who argues that the
importance of the early years has been over-stated recently, particularly in the popular press.
2 The literature on the e¤ects of child care costs and other characteristics on women’s employ-
ment has been reviewed recently by Anderson and Levine (2000) and Han and Waldfogel
(2001). For recent reviews of the literature on child care utilization decisions, see Chaplin et al.
(1999) and Han (1999).
3 See also Larner (1996) and Leibowitz (1996) for useful discussions of this issue.
4 As we have seen earlier, there is no consensus on how to define quality of child care. Child care
advocates tend to point to structural features of child care programs such as the group size,
child-sta¤ ratio, and health and safety requirements, while parents tend to look for a caregiver
who is warm and sensitive, and conveniently located. Researchers try to measure both types of
characteristics, as well as continuity and stability of care.
5 For recent evidence on this point, see Burchinal et al. (1998) and the NICHD Early Child Care
Research Network (1998).
6 Timing within the first year may matter too (Baydar and Brooks-Gunn 1991; Han et al. (2001),
Waldfogel et al. 2000).
7 See also recent reviews by Barnett (1995), Crane (ed.) (1998), Ramey and Ramey (1998a and
2000).
8 Unfortunately, not all programs tracked parental outcomes, and the few that did tended to
track outcomes for mothers only. Thus, we do not know very much about the e¤ectiveness of
early childhood child care interventions in changing the behavior of fathers.
9 Early Head Start is an early intervention program designed to enhance child development
during the first three years of life through services such as home visiting and child care. The
first 68 local programs were funded in 1995, and by 1998, more than 290 programs had been
funded. Seventeen of these programs are participating in a national evaluation by Mathema-
tica Policy Research and the Center for Young Children and Families at Columbia University
Teachers College.
10 Sure Start is an area-based initiative and allows areas to choose their own models of early
intervention so long as they meet minimum program requirements (one home visit shortly
after birth, another visit at 18–24 months, and at least a half-time day care place at age 3) and
overall program goals (to improve social and emotional development, improve health, improve
the ability to learn, and strengthen families and communities). A total of 60 ‘‘trailblazer’’ areas
were funded in 1999, and it is expected that there will be some 250 programs by the year 2002,
serving about 150,000 children or 5% of British children age 0–4.
11 Strikingly, there has been little or no research on the e¤ects of paternal employment in the pre-
school years on child outcomes. Important exceptions are Ermisch and Francesconi (2000) in
the U.K. and Ruhm (2000) in the U.S. Yet, it seems reasonable to think that early paternal
employment may have direct e¤ects on child outcomes. There may also be interactive e¤ects of
maternal and paternal employment (Han et al. 2001).
12 There is also one recent longitudinal study of maternal employment e¤ects in New Zealand
(Horwood and Fergusson, 1999); however, because this study did not separate out the e¤ects of
early maternal employment and because few of its sample members worked before the child’s 3rd
birthday, its results are not readily comparable to the U.S. or British studies discussed above.
13 Early results from the NICHD study of early child care, which is following a total of 1364
children from 10 sites across the U.S., have been reported by the NICHD Early Child Care
Research Network (1996, 1997, 1998, 1999, and 2000).
14 In the 1991–1992 British General Household Survey, 46% of families with a child under the
age of one used some form of non-parental child care, with about 16% using unpaid informal
care, 10% using nurseries, and 20% using other forms of paid care. The share of infants in care
has probably risen a good deal since then.

References
Anderson P, Levine P (2000) Child Care and Mother’s Employment Decisions. In: Card D, Blank
R (eds.) Findings Jobs: Work and Welfare Reform. Russell Sage Foundation, New York
Child care, women’s employment, and child outcomes 545

Andersson BE (1989) E¤ects of Public Day Care: A Longitudinal Study. Child Development
60:857–867
Andersson BE (1992) E¤ects of Day-Care on Cognitive and Socioemotional Competence of
Thirteen-Year-Old Swedish Schoolchildren. Child Development 63:20–36
Ball C (ed.) (1994) Start Right: The Importance of Early Learning. Royal Society of the Arts,
London
Barnett WS (1995) Long-Term E¤ects of Early Childhood Programs on Cognitive and School
Outcomes. The Future of Children 5(3):25–50
Bates JE, Marvinney D, Kelly T, Dodge KA, Bennett DS, Pettit GS (1994) Child-Care History
and Kindergarten Adjustment. Developmental Psychology 30:690–700
Baydar N, Brooks-Gunn J (1991) E¤ects of Maternal Employment and Child Care Arrangements
in Infancy on Preschoolers’ Cognitive and Behavioral Outcomes: Evidence from the Children
of the NLSY. Developmental Psychology 27:918–931
Belsky J (1988) The ‘E¤ects’ of Infant Day Care Reconsidered. Early Childhood Research
Quarterly 3:235–272
Belsky J, Eggebeen D (1991) Early and Extensive Maternal Employment/Child Care and 4–6
Year Olds Socioemotional Development: Children of the National Longitudinal Survey of
Youth. Journal of Marriage and the Family 53:1083–1099
Blau DM (1997) The Production of Quality in Child Care Centers. Journal of Human Resources
32(Spring):354–387
Blau DM, Hagy AP (1998) The Demand for Quality in Child Care. Journal of Political Economy
106(1):104–146
Blau FD, Grossberg AJ (1992) Maternal Labor Supply and Children’s Cognitive Development.
Review of Economics and Statistics: 474–481
Boocock SS (1995) Early Childhood Programs in Other Nations: Goals and Outcomes. The
Future of Children 5(3):94–114
Broberg AG, Hwang CP, Wessels H, Lamb ME (1997) E¤ects of Day Care on the Development
of Cognitive Abilities in 8-Year Olds: A Longitudinal Study. Developmental Psychology
33(1):62–69
Broberg AG, Hwang CP, Lamb ME, Ketterlinus RD (1989) Child Care E¤ects on Socioemotional
and Intellectual Competence in Swedish Preschoolers. In: Lande JS, Scarr S, Gunzenhauser N
(eds.) Caring for Children: Challenge for America. Erlbaum, Hillsdale, NJ
Brooks-Gunn J, Liaw F, Klebanov P (1992) E¤ects of Early Intervention on Cognitive Function
of Low Birth Weight Preterm Infants. The Journal of Pediatrics 120(3):35–358
Brooks-Gunn J, Klebanov PK, Liaw F, Spiker D (1993) Enhancing the Development of Low-
Birthweight, Premature Infants: Changes in Cognition and Behavior over the First Three
Years. Child Development 64:736–753
Brostrom S, Vilien K (1998) Early Childhood Education Research in Denmark. In: David T
(ed.) Research Early Childhood Education: European Perspectives. Paul Chapman Publishing,
London
Bruer JT (1999) The Myth of the First Three Years: A New Understanding of Early Brain Devel-
opment and Lifelong Learning. Free Press, New York
Burchinal MR, Roberts JE, Riggins, Jr. R, Zeisel SA, Neebe E, Bryant D (1998) Relating Quality
of Center Child Care to Early Cognitive and Language Development Longitudinally. Paper
presented at the International Conference on Infant Studies, Altanta, Georgia
Carnegie Task Force on Meeting the Needs of Young Children (1994) Starting Points: Meeting
the Needs of Our Youngest Children. Carnegie Corporation, New York
Caughy MO, DiPietro J, Strobino D (1994) Day-Care Participation as a Protective Factor in the
Cognitive Development of Low-Income Children. Child Development 65:457–471
Chaplin DD, Robins PK, Ho¤erth SL, Wissoker DA, Fronstin P (1999) The Price Elasticity
of Child Care Demand: A Sensitivity Analysis. Mimeo, The Urban Institute, Washington,
D.C.
Clarke-Stewart KA (1988) ‘The ‘E¤ects’ of Infant Day Care Reconsidered’ Reconsidered: Risks
for Parents, Children, and Researchers. Early Childhood Research Quarterly 3:293–318
Cochran MS, Gunnarson L (1985) A Follow-Up Study of Group Day Care and Family-Based
Childrearing Patterns. Journal of Marriage and the Family 47:297–309
Crane J (ed.) (1998) Social Programs That Work. Russell Sage Foundation, New York
Currie J, Thomas D (1995) Does Head Start Make a Di¤erence? American Economic Review
85(3):341–364
546 J. Waldfogel

Currie J, Thomas D (1996a) Head Start and Cognition Among Latino Children. Mimeo, Uni-
versity of California at Los Angeles, Los Angeles, CA
Currie J, Thomas D (1996b) Could Subsequent School Quality A¤ect the Long Term Gains
from Head Start? Working Paper, National Bureau of Economic Research, Cambridge,
MA
Desai S, Chase-Lansdale L, Michael R (1989) Mother or Market?: E¤ects of Maternal Employ-
ment on Cognitive Development of Four Year Old Children. Demography 26:545–561
Ermisch J, Francesconi M (2000) The E¤ects of Parents’ Working on Children’s Educational
Attainment ISER. Working Papers. Paper 2000-28, University of Essay, Colchester
Feinstein L, Robertson D, Symons J (1998) Pre-School Education and Attainment in the NCDS
and BCS. Education Economics 7(3):209–234
Field T (1991) Quality Infant Day-Care and Grade School Behavior and Performance. Child
Development 62:863–870
Francesconi M, Ermisch J (1998a) Mothers’ Behaviour and Children’s Achievement. Working
Paper 98-3, Institute for Social and Economic Research, University of Essex
Francesconi M, Ermisch J (1998b) Mothers’ Employment, Lone Motherhood, and Children’s
Achievement as Young Adults. Working Paper 98-9, Institute for Social and Economic
Research, University of Essex
Glass N (1999) Sure Start: The Development of an Early Intervention Programme for Young
Children in the United Kingdom. Children and Society 13:257–264
Gormley WT, Jr. (1995) Everybody’s Children: Child Care as a Public Problem. The Brookings
Institution, Washington, D.C.
Greenstein TN (1995) Are the ‘Most Advantaged’ Children Truly Disadvantaged by Early
Maternal Employment? E¤ects on Child Cognitive Outcomes. Journal of Family Issues
16(2):149–169
Han WJ (1999) Child Care Choices among Employed Mothers with Preschool Children. Mimeo,
Columbia University School of Social Work, New York
Han W, Waldfogel J (in press) Child Care Costs, Regulations, and the Employment of Married
and Unmarried Mothers. Social Science Quarterly (forthcoming)
Han W, Waldfogel J, Brooks-Gunn J (2001) The E¤ects of Early Maternal Employment on
Later Cognitive and Behavioral Outcomes. Journal of Marriage and the Family 63(2):336–
354
Harvey E (1999) Short-Term and Long-Term E¤ects of Early Parental Employment on Children
of the National Longitudinal Survey of Youth. Developmental Psychology 35(2):445–459
Haskins R (1985) Public School Aggression Among Children with Varying Day-Care Experience.
Child Development 56:689–703
Hayes CD, Palmer JL, Zaslow MJ (1990) Who Cares for America’s Children? Child Care Policy
for the 1990s. National Academy Press, Washington, D.C.
Hayes N (1998) Early Education Research in Ireland. In: David T (ed.) Research Early Childhood
Education: European Perspectives. Paul Chapman Publishing, London
Helburn SW (ed.) (1995) Cost, Quality, and Child Outcomes in Child Care Centers: Technical Re-
port. Department of Economics, Center for Research in Economic and Social Policy, Denver,
University of Colorado at Denver
Ho¤erth SL (1999) Child Care in the First Three Years of Life and Preschoolers’ Language and
Behavior. Mimeo, University of Michigan
Horwood LJ, Fergusson DM (1999) A Longitudinal Study of Maternal Labour Force Par-
ticipation and Child Academic Achievement. Journal of Child Psychology and Psychiatry
40(7):1013–1024
Howes C (1988) Can the Age of Entry into Child Care and the Quality of Child Care Predict
Adjustment in Kindergarten? Developmental Psychology (1990) 26(2):292–303
Howes C (1990) Relations Between Early Child Care and Schooling. Developmental Psychology
24(1):53–57
Hwang P (1990) Day Care in Sweden. In: Moss P, Melhuish E (eds.) Current Issues in Day Care
for Young Children. HMSO, London
Joshi H, Verropoulou G (2000) Maternal Employment and Child Outcomes: Analysis of Two Birth
Cohort Studies. The Smith Institute, London
Karoly LA, Greenwood PW, Everingham SS, Hoube J, Kilburn MR, Rydell CP, Sanders M,
Chiesa J (1998) Investing in Our Children: What We Know and Don’t Know about the Costs
and Benefits of Early Childhood Interventions. RAND, Santa Monica, CA
Child care, women’s employment, and child outcomes 547

Kellaghan T, Greaney BJ (1993) The Educational Development of Students Following Partici-

pation in a Preschool Programme in a Disadvantaged Area. Educational Research Centre,
Dublin
Larner M (1996) Parents’ Perspectives on Quality in Early Care and Education. In: Kagan SL,
Cohen NE (eds.) Reinventing Early Care and Education: A Vision for a Quality System. Jossey-
Bass Publishers, San Francisco, California
Lee V, Brooks-Gunn J, Schnur E, Liaw F-R (1990) Are Head Start E¤ects Sustained? A Longi-
tudinal Follow-Up Comparison of Disadvantaged Children Attending Head Start, No Pre-
school, and Other Preschool Programs. Child Development 61:495–507
Leibowitz A (1996) Child Care: Private Cost or Public Responsibility. In: Fuchs VR (ed.) Indi-
vidual and Social Responsibility: Child Care, Education, Medical Care, and Long-Term Care in
America. University of Chicago Press, Chicago, Illinois
Lynch L (2000) Trends in and Consequences of Human Asset Development. In: Danziger S,
Waldfogel J (eds.) Securing the Future: Investing in Children from Birth to College. Russell
Sage Foundation, New York
McGurk H, Kaplan M, Hennessy E, Moss P (1993) Controversy, Theory, and Social Context in
Contemporary Day Care Research. Journal of Child Psychology and Psychiatry 34:3–23
McKey RH, Condelli L, Ganson H, Barrett BJ, McConkey C, Plantz MC (1985) The Impact
of Head Start on Children, Families, and Communities. U.S. Government Printing O‰ce,
Washington, D.C.
McMahan ID (1992) Public Preschool from the Age of Two: The Ecole Maternelle in France.
Young Children 47(4):22–28
Melhuish EC (1993) Pre-School Care and Education: Lessons from the 20th and the 21st Century.
International Journal of Early Years Education 1:19–32
Melhuish EC, Lloyd E, Martin S, Mooney A (1990) Type of Day Care at 18 Months: Relations with
Cognitie and Language Development. Journal of Child Pyschology and Psychiatry 31:861–870
Melhuish EC, Moss P (1991) Day Care for Young Children: International Perspectives. Tavistock/
Routledge, London
Mocan HN, Burchinal M, Morris JR, Helburn SW (1995) Models of Quality in Center Child
Care. In: Helburn SW (ed.) Cost, Quality, and Child Outcomes in Child Care Centers: Tech-
nical Report. Department of Economics, Center for Research in Economic and Social Policy,
Denver, University of Colorado
Morgan P (1996) Who Needs Parents? Institute for Economic A¤airs, London
NICHD Early Child Care Research Network (1996) Characteristics of Infant Child Care: Factors
Contributing to Positive Caregiving. Early Childhood Research Quarterly 11:269–306
NICHD Early Child Care Research Network (1997) The E¤ects of Infant Child Care on Infant-
Mother Attachment Security: Results of the NICHD Study of Early Child Care. Child
Development 68:860–879
NICHD Early Child Care Research Network (1998) Early Child Care and Self-Control, Com-
pliance, and Problem Behavior at 24 and 36 Months. Child Development 69(3):1145–1170
NICHD Early Child Care Research Network (1999) Child Outcomes When Child-Care Center
Classes Meet Recommended Standards for Quality. American Journal of Public Health
89:1072–1077
NICHD Early Child Care Research Network (2000). The Relation of Child Care to Cognitive
and Language Development: Results from the NICHD Study of Early Child Care. Child
Development 71(4):960–980
Oliver C, Smith M, Barker S (1998) E¤ectiveness of Early Interventions. Paper prepared for
the Comprehensive Spending Review: Cross-Departmental Review of Provision for Young
Children, HM Treasury, July 1998
Osborn AF, Milbank JE (1987) The E¤ects of Early Education: A Report from the Child Health
and Education Study. Clarendon Press, Oxford
Parcel TL, Menaghan EG (1994) Early Parental Work, Family Social Capital, and Early Child-
hood Outcomes. American Journal of Sociology 99(4):972–1009
Peisner-Feinberg ES, Burchinal MR, Cli¤ord RM, Culkin ML, Howes C, Kagan SL, Yazejian N,
Byler P, Rustici J, Zelazo J (1999) The Children of the Cost, Quality, and Outcomes Study
Go to School: Executive Summary. Frank Porter Graham Child Development Center, Chapel
Hill, NC
Phillips DA (ed.) (1987) Quality in Child Care: What Does Research Tell Us? National Association
for the Education of Young Children, Washington, D.C.
548 J. Waldfogel

Ramey CT, Ramey SL (1998a) Early Intervention and Early Experience. American Pyschologist
53:109–120
Ramey CT, Ramey SL (1998b) Prevention of Intellectual Disabilities: Early Interventions to
Improve Cognitive Development. Preventive Medicine 27:224–232
Ramey SL, Ramey CT (2000) The E¤ects of Early Childhood Experiences on Developmental
Competence. In: Danziger S, Waldfogel J (eds.) Securing the Future: Investing in Children
from Birth to College. Russell Sage Foundation, New York
Rayna S, Plaisance E (1998) Early Childhood Education Research in France. In: David T (ed.)
Research Early Childhood Education: European Perspectives. Paul Chapman Publishing,
London
Reynolds A (1998) The Chicago Chiild-Parent Center and Expansion Program: A Study of
Extended Early Intervention. In: Crane J (ed.) Social Programs That Work. Russell Sage
Foundation, New York
Ruhm C (2000) Parental Employment and Child Cognitive Development. Mimeo, University of
North Carolina at Greensboro
Ruopp R, Travers J, Glantz F, Coelen C (1979) Children at the Center: Summary Findings and
Their Implications. Abt Books, Cambridge, Mass
Shore R (1997) Rethinking the Brain: New Insights into Early Development. Families and Work
Institute, New York
Smith J (1994) Maternal Employment and the Young Child. Ph.D. thesis, Columbia University
Sonnenstein FL (1991) The Child Care Preferences of Parents with Young Children: How Little Is
Known. In: Hyde JS, Essex MJ (eds.) Parental Leave and Child Care: Setting a Research and
Policy Agenda. Temple University Press, Philadelphia, Penn
Sylva K, Melhuish E, Sammons P, Sirj-Blatchford I (1999) The E¤ective Provision of Pre-School
Education [EPPE] Project Technical Paper 6a: Characteristics of Pre-School Environments.
Institute of Education, University of London, London
Sylva K, Melhuish E, Sammons P, Sirj-Blatchford I (2000) The E¤ective Provision of Pre-School
Education [EPPE] Project: A Longitudinal Study Funded by the DfEE (1997–2003). Paper
prepared for The House of Commons Education Committee Enquiry into Early Years Edu-
cation, June 2000
Sylva K, Wiltshire J (1993) The Impact of Early Learning on Children’s Later Development: A
Review Prepared for the RSA Enquiry ‘Start Right’. European Early Childhood Education
Research Journal 1(1):17–40
Vandell D, Henderson VK, Wilson KS (1988) A Longitudinal Study of Children with Day-Care
Experiences of Varying Quality. Child Development 59:1286–1292
Vandell D, Ramanan J (1992) E¤ects of Early and Recent Maternal Employment on Children
from Low-Income Families. Child Development 63:938–949
Waldfogel J, Han W, Brooks-Gunn J (2000) Early Maternal Employment and Child Outcomes: A
Longitudinal Analysis of Children from the NLSY. Mimeo, Columbia University
Zoritch B, Roberts I. The Health and Welfare E¤ects of Preschool Daycare: A Systematic Review
of Randomised Controlled Trials. Institute of Child Health, London (forthcoming)