oversight

Welfare Reform: Implications of Increased Work Participation for Child Care

Published by the Government Accountability Office on 1997-05-29.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                 United States General Accounting Office

GAO              Report to the Ranking Minority Member,
                 Subcommittee on Children and Families,
                 Committee on Labor and Human
                 Resources, U.S. Senate

May 1997
                 WELFARE REFORM
                 Implications of
                 Increased Work
                 Participation for Child
                 Care




GAO/HEHS-97-75
      United States
GAO   General Accounting Office
      Washington, D.C. 20548

      Health, Education, and
      Human Services Division

      B-270237

      May 29, 1997

      The Honorable Christopher Dodd
      Ranking Minority Member
      Subcommittee on Children and Families
      Committee on Labor and Human Resources
      United States Senate

      Dear Senator Dodd:

      In August 1996, the federal government made major changes to the
      nation’s welfare system when the Personal Responsibility and Work
      Opportunity Reconciliation Act of 1996 was enacted into law. The act
      abolished the Aid to Families With Dependent Children (AFDC) program,
      which in fiscal year 1996 spent over $20 billion providing cash assistance
      to more than 4.6 million families with about 8.6 million children. The new
      law created block grants to states for Temporary Assistance for Needy
      Families that have more stringent requirements for welfare parents to
      obtain jobs in return for their benefits than AFDC did. The new law requires
      that at least 25 percent of a state’s adult welfare caseload be working or
      participating in work-related activities in fiscal year 1997, increasing to
      50 percent by fiscal year 2002. To comply with these new work
      requirements, significantly more welfare parents are likely to need child
      care. State and local administrators are beginning to examine whether
      their current supply of child care will be sufficient to meet the increased
      demand for care—especially for particular groups of children. Data about
      the states’ child care supply will be an important tool for states in helping
      welfare parents successfully make the transition to work. Given this, you
      requested that we (1) measure the extent to which the current supply of
      child care will be sufficient to meet the anticipated demand for child care
      under the new welfare reform law and (2) identify other challenges that
      face low-income families in accessing child care.

      To accomplish our objectives, we developed a methodology for estimating
      (1) the magnitude of current demand for child care in family child care
      homes and centers, (2) the future demand under the federal welfare law,
      and (3) the extent to which the current supply of known family child care
      homes and centers is capable of meeting current and future demand. For
      purposes of this report, known care mostly consists of providers who are




      Page 1                        GAO/HEHS-97-75 Welfare Reform and Child Care Supply
B-270237




regulated by the state as well as some who are unregulated1 and listed in a
child care resource and referral agency (CCR&R) database.2 Care unknown
to the state or the CCR&Rs includes relative care, unregulated family child
care, and care provided in a child’s home by a nanny. We elaborate further
on the different types of child care later in this report.

In calculating our estimates, we held the supply of known child care
constant. While the total supply of child care, known and unknown to the
states, should eventually increase in response to an increase in demand,
our model, by holding known supply constant, presents a picture of how
much current known supply would have to increase to meet the new
demand. Thus, the gap we identify is between known supply and
anticipated demand. This gap could be filled by care known or unknown to
the states or, more likely, by both.

Ideally, we would like to have measured the total supply of child care. We
recognize the important role that other types of care play in meeting child
care demand; however, our review focused on known care because it is
the type of care for which states and localities have the most
comprehensive data. In focusing on known care, we are making no
judgment about the quality of either type of care.

In developing our information, we analyzed child care supply data and
estimated child care demand at four sites—two urban and two
nonurban—in three states. Our selected urban areas were Baltimore City,
Maryland, and Chicago, Illinois; our nonurban sites were Benton and Linn
counties in Oregon. We selected these sites not only because they
provided a mix of urban and nonurban areas, but also because
comprehensive child care supply data were available from local CCR&Rs.
Other factors considered in our final selection were geographic diversity
and differences in the extent to which sites regulated their child care
providers.3

1
 Regulated care is offered by providers whom the state requires to obtain a license or become
registered with the state; unregulated care is offered by providers whom the state does not require to
register to provide such services. The three states we reviewed require unregulated providers to meet
some requirements if they are caring for children whose child care costs are paid by the state.
Typically, the provider signs a form or checklist to certify that it meets or will follow certain basic
safety and health requirements.
2
 CCR&Rs help match parents looking for child care with providers who can serve their specific child
care needs. Typically, these agencies are funded by state or local child care agencies, by private
employers, and through charitable contributions. In addition to helping parents find care, CCR&Rs
perform other services, including recruiting and training providers, helping states administer child care
subsidy programs, and maintaining a current and comprehensive database of an area’s child care
supply.
3
 See app. I for further discussion of our methodology.


Page 2                                   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                   B-270237




                   As states implement the new welfare reform legislation and are required to
Results in Brief   move larger percentages of their caseloads into work-related activities,
                   greater numbers of welfare recipients are likely to need child care.
                   Consequently, the gap that exists between the current supply of known
                   child care and child care demand is likely to grow, with disproportionately
                   larger gaps for infants and school-aged children. These gaps will have to
                   be addressed through growth in the supply of known child care; care
                   unknown to the states; or, more likely, both. If supply of known child care
                   does not increase, states may have to rely more on care for which they
                   have little information. Thus, the assistance they can provide to welfare
                   parents in locating care may be more limited.

                   State and local officials in the four cities and counties we reviewed
                   regarded their current supply of known child care as inadequate for
                   meeting even the demand they currently face for children in certain age
                   groups, particularly for low-income populations in three of the areas
                   reviewed. Unless the supply of known child care for certain age groups at
                   these sites is increased, the gap between supply of known care and
                   anticipated demand is likely to become even greater as welfare reform is
                   fully implemented. For example, we estimated that the supply of known
                   child care in Chicago would be sufficient to meet just 14 percent of the
                   demand for infant care that will probably exist by the end of fiscal year
                   1997—1 year after enactment of the welfare reform legislation. Without
                   any increase, by the year 2002, when states will be required to have
                   achieved welfare work participation rates of 50 percent, the known supply
                   could meet only about 12 percent of the estimated demand for infant care
                   and even less in the poorest areas of Chicago. Thus, we estimated that, by
                   the end of fiscal year 1997, the demand for infant care could exceed the
                   known supply by about 20,000 spaces; by fiscal year 2002, this number
                   could increase to almost 24,000.

                   Issues other than gaps between supply and demand that could also affect
                   low-income families’ access to care include the price of care, the
                   availability of nonstandard-hour care, transportation issues, and the
                   availability of quality care. For example, our work shows that child care
                   consumes a high percentage of poor families’ income. In Benton County,
                   Oregon, infant care at a child care center consumes more than 20 percent
                   of the median household income for a poor family. Another critical issue
                   facing poor families is that many welfare parents are likely to obtain work
                   at low-skill jobs that operate on nonstandard schedules, such as janitor or
                   cashier. However, many of the known providers at the sites reviewed did
                   not offer child care at nonstandard work hours—hours outside the



                   Page 3                       GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                           B-270237




                           traditional “9 to 5” work schedule. The number of providers who offered
                           this type of care ranged between 12 and 35 percent of the total number of
                           known providers in the four child care markets we reviewed.



Background

Welfare Reform and Child   Before the Personal Responsibility and Work Opportunity Reconciliation
Care Demand                Act of 1996, the Congress had enacted changes to the nation’s welfare
                           system in 1988 by passing the Family Support Act (FSA). This law created
                           the Job Opportunities and Basic Skills Training (JOBS) program, which
                           expanded upon previous programs designed to help families on welfare
                           obtain education, training, and work experience to become self-sufficient.
                           The Department of Health and Human Services (HHS) is the agency at the
                           federal level that was responsible for JOBS program administration and
                           oversight.

                           Recognizing the importance of child care to this effort, the Congress
                           provided for child care subsidies to welfare parents participating in JOBS
                           activities and to those who had recently moved from welfare to work. In
                           fiscal year 1996, the Congress appropriated $954 million to the states
                           through the child care programs established by FSA to help parents pay for
                           child care.4 States also used the Child Care Development Block Grant
                           (CCDBG) as another source of funding to pay for the child care of JOBS
                           participants. CCDBG was created by the Omnibus Reconciliation Act of
                           1990; in fiscal 1996, the Congress appropriated $935 million to the states
                           through CCDBG.

                           FSA required states to have specific percentages of their welfare caseloads
                           participate in JOBS activities, starting at 7 percent in fiscal year 1991 and
                           rising to 20 percent by the end of fiscal year 1995. However, these
                           participation requirements were not applicable to the states’ entire welfare
                           caseload. States were permitted to exempt from JOBS activities welfare
                           clients who were already working 30 hours or more a week; ill or
                           incapacitated in some way; full-time students in elementary, vocational, or
                           high schools; children under the age of 16; or caring for a child under the


                           4
                            This amount is the federal appropriation for fiscal year 1996; it does not include the amount of state
                           dollars used. FSA created two sources of child care funding to be used by the states —AFDC child
                           care and transitional child care. Both were matched, open-ended entitlements, which meant that, in
                           order for a state to use the federal funds, the state had to first match the federal amount it needed with
                           state dollars.



                           Page 4                                   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                      B-270237




                      age of 3.5 Individuals who had not been exempted were required to
                      participate in 20 hours of JOBS activities per week, on average. If child care
                      was unavailable or resources were unavailable to pay for care, states
                      could either exempt welfare families or limit their participation in JOBS
                      activities. Because of these and other provisions, as well as the amount of
                      resources states devoted to their JOBS programs, the number of welfare
                      families participating in JOBS nationwide was limited—about 13 percent of
                      the entire caseload in any given month in fiscal year 1994. This low level of
                      participation limited demand for certain types of care that were more
                      difficult to find, such as infant care and care during nonstandard work
                      hours.6

                      Compared with the former welfare provisions, the Personal Responsibility
                      and Work Opportunity Reconciliation Act incorporates more stringent
                      work participation provisions and requires that a larger proportion of
                      welfare parents obtain work or participate in work-related activities. For
                      example, the new law requires states to move at least 25 percent of their
                      welfare families into work or work-related activities by fiscal year 1997
                      and at least 50 percent by fiscal year 2002. All welfare parents are required
                      to participate in these activities, but states have the option to exempt
                      single parents who are caring for a child up to 1 year old. States are also
                      permitted to lower the age-of-child exemption. Michigan, for example,
                      requires parents with children 3 months old or older to obtain work or
                      engage in work-related activities. Under the new law, more mothers are
                      likely to need child care, particularly for very young children.


Types of Child Care   Parents can choose from three types of child care settings: in-home care,
Settings              where a child is cared for in the child’s home; family care, where the child
                      is cared for in the home of a provider; and center care, where a child is
                      cared for in a nonresidential setting. In-home care, such as that provided
                      by au pairs or nannies, is usually provided for the child or children of one
                      family that resides in the home. Family child care, on the other hand, is
                      provided to a small number of unrelated children—typically fewer than
                      six—in the provider’s home. Child care centers, also known as nursery
                      schools or preschools, are nonresidential facilities that are able to care for
                      much larger numbers of children—on average, about 60. Centers are
                      located in a variety of places, including churches, schools, and businesses.


                      5
                       States were permitted to change the age-of-child exemption so that parents with children 1 year old or
                      older could be required to participate in JOBS activities. According to state plans submitted to HHS for
                      fiscal years 1994-1996, 13 states had planned to use this option.
                      6
                       Welfare to Work: Child Care Assistance Limited; Welfare Reform May Expand Needs
                      (GAO/HEHS-95-220, Sept. 21, 1995), pp. 4-6.


                      Page 5                                  GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                       B-270237




                       Additionally, care can be provided in family child care or in-home settings
                       by someone related to the child other than the parents, such as a
                       grandparent or an aunt. This care is commonly called relative care.


Child Care Providers   Child care supply in a local market consists of providers who are known
                       and unknown to the states. Known providers are mostly those who are
                       regulated by the state but also include some who are unregulated.
                       Providers that are regulated by the state are required to meet certain
                       standards for operating that unregulated providers may or may not meet.
                       Such standards specify, for example, the number of smoke detectors a
                       provider needs to have, the maximum number of children per staff person
                       allowed, or that children are required to be immunized before coming into
                       a provider’s care. These standards are established by the state or local
                       government, and compliance with them is monitored periodically by the
                       governing entity.

                       Most center care in states is regulated, although some states exempt
                       centers from regulation if they are sponsored by a religious group or a
                       government entity or are regulated by another government body, such as a
                       local education department. Much family child care is unregulated: One
                       study has estimated that between 10 and 18 percent of family child care
                       homes are regulated.7 In-home care and care by relatives is almost never
                       regulated by states.

                       States and localities maintain data about providers that can help parents in
                       choosing a provider and states and localities in determining the extent of
                       supply. Obtained from either CCR&Rs or state licensing offices, these data
                       include, for example, the total number of known providers at a given time,
                       whether providers are centers or family child care homes, where providers
                       are located, the number and ages of children they serve, how much they
                       charge, and their hours of operation. With these data, states have a tool to
                       facilitate matching child care needs of individual welfare families with the
                       services offered by certain providers. Helping parents find providers is
                       important given the difficulty welfare parents may have in negotiating the
                       child care market on their own, the barrier that finding child care can
                       become to welfare parents’ labor force participation, and the states’ own
                       incentive to move their welfare caseload into the workforce to meet
                       requirements of the new federal law. In addition, information about the



                       7
                        Willer, B., Hofferth, S., Kisker, E.E., and others, The Demand and Supply of Child Care in 1990
                       (Washington, D.C.: National Association for the Education of Young Children, 1991).



                       Page 6                                   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                          B-270237




                          supply of known child care providers allows states to target groups of
                          providers for communitywide supply-building efforts, if needed.


Factors Influencing the   The child care setting parents use is frequently related to the age of the
Choice of Child Care      child and the employment status of the mother. Many mothers with
Setting                   children under the age of 2 who work full time place their children in the
                          care of a family child care provider as opposed to a center or in-home
                          setting. For example, in 1993, 40 percent of children under 1 year of age
                          whose mothers worked were cared for in a provider’s home; 38 percent of
                          children between 1 and 2 were also cared for in this setting. As children
                          reach preschool age, care in organized facilities, such as child care centers
                          and nursery schools, becomes more prominent, although the use of family
                          child care is still significant. Of children who were 3 to 4 years old and had
                          working mothers in 1993, approximately 37 percent were cared for in
                          centers. For those who were 4 to 5 years of age and whose mothers were
                          working, almost 42 percent were cared for in centers.8

                          Other important factors, such as family income, marital status, and the
                          number of days and hours of the day worked, also influence the choice of
                          child care setting. For example, low-income families whose annual income
                          falls below $15,000 generally rely more on relative care and less on
                          center-based care than do nonpoor families.9 Low-income mothers who
                          are single and employed also rely heavily on relative care, although they
                          also make significant use of family child care homes and centers.10
                          Relative care is also used frequently by families whose jobs require them
                          to work nonstandard hours.




                          8
                          GAO analysis of the Survey of Income and Program Participation, U.S. Bureau of the Census, U.S.
                          Department of Commerce, fall 1993.
                          9
                            National Research Council, Child Care for Low-Income Families (Washington, D.C.: National
                          Academy Press, 1995), p. 7.
                          10
                           Thirty percent of the children of low-income mothers who are single and employed are with relatives,
                          27 percent are in centers, and 21 percent are in family child care homes (Child Care for Low-Income
                          Families, p. 6).



                          Page 7                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                    B-270237




                    For a number of years, the child care literature has documented the
Greater Need for    difficulty in finding care for certain age groups of children and specific
Infant and          types of child care, including care for infants and school-aged children,
School-Aged Care    children with disabilities, and children during nonstandard work hours.
                    Our earlier work found that, because of shortages in child care for infants,
Anticipated Under   school-aged children, sick children, and children with special needs as
Welfare Reform      well as shortages of care during nonstandard work hours, state and county
                    administrators had difficulty serving the child care needs of welfare
                    parents who were participating in the JOBS program.11

                    CCR&R  staff, as well as state and local officials at the four sites we visited,
                    said that finding some of these types of care is still difficult. Officials at all
                    sites were concerned that the new federal welfare act or their own state
                    welfare initiatives might increase demand for certain types of care and
                    further exacerbate low-income parents’ problems in finding care. We
                    estimated that the current supply of known care in the four cities and
                    counties we reviewed can meet less of the demand for infant or
                    school-aged care than it can for preschool care (see table 1). We projected
                    that this gap will grow as higher work participation rates are required
                    under the new welfare act, assuming no growth in known supply. In Linn
                    County, Oregon, for example, we estimated that the supply of known child
                    care is sufficient to meet 45 percent of the current demand for school-aged
                    care. By the end of fiscal year 1997, when 25 percent of the state’s welfare
                    recipients will be required to participate in work or a work-related activity,
                    known supply could be sufficient to meet 43 percent of school-aged
                    demand, assuming there is no growth in that supply. In fiscal year 1999,
                    when 35 percent of the state’s welfare caseload will be required to
                    participate, known supply could be sufficient to meet 42 percent of the
                    demand; in fiscal year 2002, when participation rates of 50 percent are
                    implemented, known supply could be sufficient to meet 40 percent of the
                    demand.




                    11
                      GAO/HEHS-95-220, Sept. 21, 1995.



                    Page 8                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                       B-270237




Table 1: Estimates of the Percentage
of Child Care Demand That Could Be                              Percentage of Percentage of demand that could be met
Met by Currently Known Supply for                         current demand that      by currently known supply under
Various Age Groups, 1996-2002                                 could be met by     welfare reform work requirements
                                                              currently known           25%           35%           50%
                                                                        supply participation participation participation
                                       Baltimore City
                                       Infant                              37            33            32            30
                                       Preschool                          144           130           125           118
                                       School-aged                         49            43            41            38
                                       Chicago
                                       Infant                              16            14            13            12
                                       Preschool                           75            68            65            62
                                       School-aged                         23            20            19            17
                                       Benton County
                                       Infant                              67            64            63            61
                                       Preschool                           92            90            89            87
                                       School-aged                         66            64            63            62
                                       Linn County
                                       Infant                              44            40            39            38
                                       Preschool                           74            71            69            68
                                       School-aged                         45            43            42            40

                                       At all of our sites, CCR&R staff were the least concerned about the
                                       availability of preschool care, as compared with that for other age groups,
                                       because they believed the supply of known preschool care is the most
                                       adequate to meet demand. In Baltimore City, for example, we estimated
                                       that the supply of known preschool care exceeds the current demand as
                                       well as estimated future demand under the new welfare act, as shown in
                                       table 1. Given the city’s current excess known supply for this age group,
                                       CCR&R staff in Baltimore City said they are not encouraging prospective
                                       providers to offer care for preschoolers.

                                       While examining the percentage of demand that known supply is capable
                                       of meeting is a useful way to present a picture of child care at various
                                       sites, it masks the size of the problem at some sites. For example, in
                                       percentage terms, the supply of known infant care in Baltimore City could
                                       be sufficient to meet about 37 percent of the demand for such care; in
                                       absolute terms, that supply could leave an unmet demand for care of over
                                       3,000 infants, as shown in table 2. Similarly, while known supply could be
                                       sufficient to meet about 75 percent of the demand for preschool care in




                                       Page 9                        GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                    B-270237




                                    Chicago, it could leave the parents of more than 13,000 preschool children
                                    looking elsewhere for care.

Table 2: Estimated Gaps Between
Known Supply and Demand for Child                                                            Gaps between known supply and
Care for Various Age Groups,                                           Current gap          demand under welfare reform work
1996-2002                                                           between known                     requirements
                                                                        supply and        25%           35%           50%
                                                                          demand participation participation participation
                                    Baltimore City
                                    Infant                                       3,369            4,037             4,304            4,704
                                               a                                       a                 a                a                 a
                                    Preschool
                                    School-aged                                  6,115            7,901             8,615            9,687
                                    Chicago
                                    Infant                                     17,046            20,402           21,744           23,757
                                    Preschool                                  13,450            19,247           21,566           25,045
                                    School-aged                                26,393            31,590           33,669           36,787
                                    Benton County
                                    Infant                                         147              171               181                 196
                                    Preschool                                      100              136               150                 172
                                    School-aged                                    355              389               403                 424
                                    Linn County
                                    Infant                                         388              443               465                 498
                                    Preschool                                      470              561               597                 652
                                    School-aged                                    828              920               957            1,012
                                    a
                                     No numbers are shown for preschool children in Baltimore City because we estimated that
                                    supply for preschool care currently exceeds demand and will do so under welfare reform.




                                    Areas with the lowest average household income will probably be most
Need for Infant and                 affected by welfare reform. As shown in table 3, the largest gaps between
School-Aged Care                    known supply and demand in the poor areas12 of the selected sites exist
Greatest in Poor                    for infants, school-aged children, or both. For example, in poor areas of
                                    Chicago, currently known supply is sufficient to meet 61 percent of the
Areas                               demand for preschool care, compared with 11 percent and 30 percent of
                                    the demand for infant and school-aged care, respectively. A similar



                                    12
                                     In general, we defined poor areas as those census tracts with median household incomes below
                                    $27,750, or slightly over 200 percent of poverty for a family of four. We recognize that child care
                                    markets are neither defined by census tracts alone nor completely segregated so that poor parents
                                    purchase care only in poor areas and nonpoor parents in nonpoor areas. Research does suggest,
                                    however, that parents prefer to use providers close to their home.



                                    Page 10                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                         B-270237




                                         relationship between these types of care exists in nonpoor areas of
                                         Chicago as well.


Table 3: Estimates of the Percentage of Child Care Demand That Could Be Met by Currently Known Supply for Poor and
Nonpoor Areas, 1996-2002
                               Percentage of current demand that    Percentage of demand that could be met by currently
                                could be met by currently known    known supply in poor areas under welfare reform work
                                             supply                                     requirement
                               Nonpoor areas         Poor areas   25% participation   35% participation   50% participation
Baltimore City
Infant                                     48                32                  27                 25                  23
Preschool                                 237               109                  94                 89                  83
School-aged                                75                36                  29                 27                  25
Chicago
Infant                                     22                11                   8                  7                    7
Preschool                                 105                61                  53                 50                  46
School-aged                                21                30                  24                 22                  20
Benton County
Infant                                     62                69                  61                 58                  54
Preschool                                  98                80                  75                 73                  71
School-aged                                66                60                  54                 52                  50
Linn County
Infant                                     48                34                  29                 28                  26
Preschool                                  92                44                  41                 40                  38
School-aged                                49                35                  31                 30                  28

                                         While currently in both poor and nonpoor areas the gap between demand
                                         and known supply is greatest for both infant and school-aged care, this
                                         condition could worsen in poor areas of our sites as the welfare reform
                                         legislation is implemented. Given that families on welfare generally live in
                                         poor areas, the increase in demand for child care resulting from welfare
                                         reform will probably be greater in poor areas than in nonpoor areas. Thus,
                                         for example, while the supply of known school-aged care in the poor areas
                                         of Baltimore City is sufficient to meet about 36 percent of current demand,
                                         assuming no growth in the known supply, the percentage could decrease
                                         to about 25 percent in fiscal year 2002, when 50 percent of the welfare
                                         caseload is required to participate in work or work-related activities.

                                         As previously discussed, some of the largest current and future gaps in the
                                         known supply could exist for infants at the four sites. In some instances,




                                         Page 11                         GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                           B-270237




                           these gaps are even greater in poor areas than at the site as a whole. For
                           example, although we estimated that the known supply for all of Chicago
                           is sufficient to meet about 16 percent of current demand for infant care
                           (see table 1), the known supply in poor areas of Chicago is capable of
                           meeting only 11 percent of demand (see table 3). As implementation of the
                           welfare reform legislation progresses, this figure could drop to 7 percent
                           in fiscal year 2002.

                           A gap also exists between the percentage of demand that is met by the
                           supply of known preschool care overall and the percentage that is met in
                           different areas. For example, in Linn County, Oregon, we estimated the
                           current known supply to be sufficient to meet 74 percent of current
                           demand for preschool care in the entire area. However, in poor areas of
                           the county, known supply could be sufficient to meet only 44 percent of
                           demand, and this figure could drop to 38 percent in fiscal year 2002 under
                           welfare reform.


                           At all of our sites, CCR&R staff or state and local officials cited other issues
Low-Income Families        that low-income families face in accessing care, including the price of
Face Other Issues in       care, the availability of nonstandard-hour care, transportation issues, and
Accessing Child Care       quality of care.13

Price of Care Consumes     At the four sites we reviewed, the price of known child care consumed a
Large Percentage of Poor   large percentage of household income for low-income families.14 As shown
Families’ Incomes          in table 4, the median price of full-time infant care as a percentage of
                           median household income ranged from 16 to 43 percent for poor families.
                           The range for full-time care for preschool children was from 14 to
                           24 percent, and for school-aged children, from 8 to 18 percent.15 These
                           percentages do not take into account the possibility that some low-income




                           13
                            In our general discussions with CCR&R staff and state and local officials about child care issues
                           affecting low-income families, we assumed low-income families to be those whose annual incomes
                           qualified them to receive child care subsidies, regardless of whether or not they received welfare.
                           Eligibility for child care subsidies differed in each state we visited.
                           14
                             Technically, it is the parents’ share of the price of care that should be measured relative to household
                           income. However, our databases contained only the total price charged by the provider. These two
                           numbers will differ to the extent that child care is subsidized by a third party.
                           15
                             For those providers that provided hourly rates instead of weekly ones, we assumed full-time care for
                           infants and preschoolers to be 45 hours per week and full-time care for school-aged children to be 25
                           hours per week.



                           Page 12                                  GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                           B-270237




                                           families may receive a child care subsidy.16 However, national survey data
                                           for 1993, which include families with and without subsidies, also show that
                                           low-income families who paid for care spent, on average, as much as
                                           18 percent of their income on child care expenses.17


Table 4: Median Weekly Price for Known Child Care in Poor and Nonpoor Areas (Absolute Values and as a Percentage of
Sites’ Median Household Income)
                                    Center care                                        Family day care
                        Poor                          Nonpoor                             Poor                            Nonpoor
                           % of median                     % of median                      % of median                           % of median
                     Price      income               Price      income                Price      income                 Price          income
Baltimore City
Infant                $154           38              $147               22              $86               21              $86                 13
Preschool            80-81           20              82-85          12-13                76               19               76                 12
School-aged          43-75         10-18             45-81            7-12           50-75            12-19            50-75                8-11
Chicago
Infant             130-153         37-43        133-155             18-21                85               24               85                 12
Preschool            81-84         23-24             80-81              11           65-80            18-23            70-80              10-11
School-aged             56           16                45                6               60               17               60                  8
Benton County
Infant                 121           23               109               13               90               17               90                 11
Preschool               85           16                88               10               79               15               90                 11
School-aged             43             8               32                4               44                8               50                  6
Linn County
Infant                  97           20               101               17               79               16               79                 13
Preschool               69           14                73               12               68               14               70                 11
School-aged             41             8               43                7               38                8               38                  6
                                           Note: We used 1990 U.S. Census data to calculate median household income for poor and
                                           nonpoor areas. We used only those providers that reported a price in the CCR&R databases in
                                           1996 to calculate the median weekly price of care.



                                           The difference in the percentage of household income that the price of
                                           known child care represents for poor and nonpoor families is almost
                                           entirely due to differing median household incomes rather than
                                           differences in the child care prices themselves. At our four sites, in most
                                           cases, price differences between poor and nonpoor areas were small and


                                           16
                                            Our previous work and that of others has shown that many eligible families do not receive child care
                                           subsidies, mostly because of state funding constraints. See Child Care: Working Poor and Welfare
                                           Recipients Face Service Gaps (GAO/HEHS-94-87, May 13, 1994).
                                           17
                                            U.S. Bureau of the Census, What Does It Cost to Mind Our Preschoolers? Current Population
                                           Reports, P70-52 (Washington, D.C.: U.S. Government Printing Office, 1995).


                                           Page 13                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
B-270237




generally less for homes than for centers.18 For example, the price of
center care for infants in poor areas of Chicago ranged from $130 to $153
and consumed between 37 and 43 percent of median household income for
poor families. The price range for infant care in nonpoor areas was $133 to
$155, but the percentage of median household income consumed was
between 18 and 21 percent for nonpoor families (see table 4).

CCR&R staff in Oregon and Chicago were surprised by the similarity in
prices of known care in poor and nonpoor areas. However, national data
show that, while many poor families may secure child care free of cost,
those poor families that do pay for care pay an amount not significantly
different from that paid by nonpoor families. Hence, “...poor families that
do pay for child care may compete against more financially able families
for child care services, and hence pay competitive prices for these
services.”19

Both CCR&R staff and state and local officials in the four cities and counties
said that the affordability of child care was a barrier for low-income
families in accessing child care. For example, Oregon CCR&R staff said that
money to buy child care, especially the more expensive infant care or care
for a child with special needs, is as much an issue for low-income families
as is building the supply of this type of care. They believe that if parents
had more money to purchase care, more providers would be willing to
offer it. Officials in Baltimore City and Chicago told us that the
affordability of child care for low-income families depends on the
subsidies they receive. In Chicago, both CCR&R staff and state officials told
us that subsidy rates are too low for some types of care and for
low-income parents who must compete with families who have more
resources to find care in economically mixed neighborhoods. Low
subsidies also contribute to high turnover for providers caring for
low-income children and create instability for both children and parents.
In Baltimore City, however, CCR&R staff believe that their subsidy rates
provide access to quality care for those families that receive them.




18
  With a few exceptions, these differences ranged from approximately $1 to $5 a week for centers and
from $0 to $3 per week for homes. However, in a few instances, price differences were larger between
the areas. For example, the median price for school-aged care in Chicago was $56 per week in poor
areas as compared with $45 per week in nonpoor areas, and the price of preschool family day care was
$79 in poor areas of Benton County and $90 in nonpoor areas.
19
  Bureau of the Census, Who’s Minding the Kids? Child Care Arrangements: Fall 1991, Survey of
Income and Program Participation (Washington, D.C.: U.S. Department of Commerce, 1994) p. 23.



Page 14                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                          B-270237




Few Providers Offer       Given their often limited education and low skill levels, many parents
Nonstandard-Hour Care     moving from welfare to work may become employed in jobs with
                          nonstandard-hour work schedules, such as jobs that have rotating shift or
                          weekend hours, for example.20, 21 These types of jobs include cashier,
                          retail salesperson, and janitor. Finding child care during nonstandard work
                          hours may prove challenging, however. Our previous work examining the
                          use of child care under the JOBS program found that state and local
                          officials were having difficulty finding child care during nonstandard hours
                          for the children of AFDC parents participating in the JOBS program.22

                          At the four sites we reviewed, fewer providers offered nonstandard-hour
                          care as compared with other types of care: The percentage of providers
                          that offered nonstandard-hour care ranged from 12 percent to 35 percent.
                          Providers that offered nonstandard-hour care were predominantly family
                          child care homes, not centers, which have significantly greater capacity
                          than homes. Appendix II provides detailed data on the number and type of
                          providers offering this type of care by site.


Transportation Issues     Transportation is another critical issue for welfare families in accessing
Affect Accessibility of   child care. As we previously reported, in a nationwide survey of local JOBS
Child Care                program officials, 23 percent stated that they could not meet the child care
                          needs of all their participants, and 77 percent of these reported that this
                          was because of transportation problems. States reported that JOBS
                          participants did not have reliable private transportation to get their
                          children to child care providers and then to get themselves to work.
                          Moreover, some communities lacked the necessary public transportation
                          to get participants where they needed to go, especially in rural areas.23

                          CCR&R staff at all four of the sites reviewed also stated that lack of
                          transportation created barriers for low-income families in obtaining child
                          care. In addition to the transportation issues cited above, CCR&R staff in
                          Chicago and Oregon said that transportation of school-aged children
                          between school and their after-school provider was a problem.


                          20
                             U.S. Congressional Research Service, Jobs for Welfare Recipients (Washington, D.C.: Library of
                          Congress, May 13, 1994), pp. 2-5.
                          21
                           Presser, H., Jobs, Family, and Gender: Determinants of Nonstandard Work Schedules Among
                          Employed Americans in 1991 (College Park, Md.: Center of Population, Gender, and Social Inequality,
                          University of Maryland, 1995).
                          22
                            GAO/HEHS-95-220, Sept. 21, 1995, p. 4.
                          23
                            GAO/HEHS-95-220, Sept. 21, 1995, p. 9.



                          Page 15                                    GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                            B-270237




States and CCR&Rs Have      Whether provided in centers or family child care homes, child care of
Concerns About Child Care   acceptable quality is care that nurtures children in a stimulating
Quality                     environment, safe from harm. Research has documented that elements in a
                            child care setting that are associated with an acceptable level of quality
                            include trained providers, small group sizes, low child-to-staff ratios, and
                            low staff turnover, to name a few. Research over many years also has
                            documented the importance of the quality of care to all aspects of a child’s
                            healthy development—physical, cognitive, emotional, and social.

                            CCR&R staff and state and local officials at all four sites told us they were
                            concerned about the quality of care that low-income families are able to
                            access. For example, CCR&R staff in Oregon said that, given the numerous
                            constraints faced by low-income families, including low wages, less
                            flexible hours, and a lack of transportation, many low-income parents
                            have limited child care choices, which decreases their chances of finding
                            care of acceptable quality. In fact, one Oregon state official believes that
                            the supply of child care will be less of a critical issue under welfare reform
                            than the quality of the care parents access. In Baltimore City, CCR&R staff
                            and state and local officials were concerned about low-income families
                            not having access to care of acceptable quality unless they had access to
                            child care subsidies with which to purchase such care.24 Chicago CCR&R
                            staff expressed similar concerns and also said that the subsidy rates for
                            some types of care in Chicago are too low to purchase child care of an
                            acceptable level of quality.


                            The availability of child care will be a key factor in the degree to which
Concluding                  states and the federal government succeed in helping welfare families
Observations                become more financially self-sufficient. On the basis of our review of four
                            sites, it seems likely that increases in the supply of child care will be
                            needed to meet the estimated increase in demand for care as welfare
                            reform is implemented. Questions remain, however, about which segments
                            of the child care market these increases will come from; how much care
                            will be needed; and whether or not states and localities can effectively
                            help increase the supply where it is most needed. The answers to these
                            questions, for our sites and other communities as well, depend on factors
                            such as the following:


                            24
                              At the time of our review in Baltimore City, Maryland, CCR&R staff and state and local officials
                            stated that many families could not get child care subsidies because of a lack of state funds and were
                            on a waiting list. However, comments provided by CCR&R staff and Maryland state officials on our
                            draft report indicated that this situation had changed. More state and federal money has become
                            available for child care subsidies, and, as a result, Maryland has eliminated its waiting list for such
                            subsidies.



                            Page 16                                  GAO/HEHS-97-75 Welfare Reform and Child Care Supply
B-270237




Proportion of demand currently being met: If states’ current supply of
child care is sufficient to meet or exceed current demand, an increase in
demand over time may not be a problem. The growth in the supply of child
care experienced by an area may match the demand. However, if the
current supply of care is quite low relative to the demand, as our estimates
showed for infant care at our selected sites, and demand increases with
welfare reform, it is possible that the growth in total supply might not be
quick enough to meet the increase in the short term.

Ability of states to affect supply: While the supply of child care is expected
to grow in response to more demand, states and communities can, for the
most part, directly reach only those providers known to the states. With
the information available from these providers, the state can target them in
its supply-building activities, if needed. Such activities could include
providing incentives or subsidies, or making regulatory changes in an
effort to directly increase the supply overall or for a particular type of
care. The bigger challenge for states and localities will be increasing the
supply of care unknown to the states, such as that provided by friends,
neighbors, and especially relatives, which is chosen by many parents to
meet their child care needs. Information on these providers, however, is
limited.

Age of children needing care: The new federal law dramatically changes
which welfare parents are required to find work. Previously, that group
consisted of parents whose youngest child was 3 years old or older; now,
the work requirement applies to the entire welfare population, except for
parents with children 1 year old or younger. Young children are primarily
cared for in a provider’s home; a significant number of these providers are
relatives, particularly in low-income families. Hence, the increase in
demand for care for very young children caused by the new law could
place the greatest strain on the supply of family child care providers and
relative care. Information about family child care providers is limited
because many are unknown to the states; furthermore, their individual
capacity typically is limited. In addition, states and local governments
generally have little information available on relative care. States’ inability
to directly expand the capacity of these types of providers could pose
significant obstacles to communities in their efforts to ensure that their
child care supply meets the needs of their welfare parents.

Price of care: The price of child care can have a positive effect on building
supply, assuming that the amount parents are able and willing to pay is
high enough to attract more providers to the market. On the other hand,



Page 17                        GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                     B-270237




                     the higher the price of child care, the less affordable care becomes for
                     low-income families, especially for those without child care subsidies.
                     Data from our sites as well as national data show that child care,
                     especially infant care, consumes a high percentage of household income
                     for poor families. As a result, child care subsidies could become critical to
                     low-income families’ ability to afford care and, as officials in some of our
                     sites stated, buy quality care. However, the extent to which states have the
                     resources to provide subsidies to greater numbers of eligible families and
                     whether or not the amount of those subsidies will be high enough to build
                     supply are not known at this time.

                     The way in which these four factors interact in each market, and the
                     extent to which states and localities can influence these factors, will affect
                     the expansion of child care supply, which is important to welfare parents
                     who are making the transition to work.


                     We obtained comments on a draft of this report from HHS and state and
Comments From HHS,   CCR&R child care officials from the four areas reviewed in this report. HHS
States, and CCR&Rs   officials said that the report’s findings reflect some of the child care issues
and Our Evaluation   they have heard across the country, such as the gap between the supply of
                     and demand for infant and school-aged child care; the current inadequacy
                     of supply that states and communities face, particularly in low-income
                     areas; and the significant portion of a low-income family’s income that
                     child care consumes. HHS officials also noted that the report reinforces
                     earlier GAO work regarding difficulties that state and county administrators
                     have had in serving the child care needs of welfare parents participating in
                     the JOBS program. They also believe that the report is a useful next step in
                     identifying the crucial role child care plays in the lives of working families.
                     HHS’ written comments appear in full in appendix III.


                     Some CCR&Rs and state officials expressed several concerns related to our
                     not including unregulated child care in the scope of our review. First, by
                     excluding such care, they said, the report understates the importance and
                     significance of unregulated care in meeting the child care needs of welfare
                     recipients. The officials went on to say that caregivers such as relatives,
                     friends, and neighbors currently meet the needs of many parents,
                     particularly low-income parents, and that these providers will be an
                     important source of supply as demand grows in response to welfare
                     reform. Although our report focuses on known family child care homes
                     and centers, we recognize the important role that other types of care play
                     in meeting child care demand. The scope of our review was limited to



                     Page 18                        GAO/HEHS-97-75 Welfare Reform and Child Care Supply
B-270237




known care only because of methodological constraints in attempting to
measure the total supply of care: As we state in the report, care both
known and unknown to the states will be important in filling the estimated
gap between known supply and future demand. In response to these
comments, we have revised the report to further acknowledge the
importance of other types of care.

Some CCR&Rs and state officials also believed that our estimates of future
demand for child care were based on an assumption that parents would
use only family child care homes or center care. In reality, our estimates of
future demand for care were based on the assumption that parents will use
all types of child care in the same proportions as they are currently using
them. Thus, we compared the supply of known family child care homes
and centers with only the demand for family child care homes and centers.

Finally, CCR&Rs and state officials believed that our discussion of the
quality of child care accessed by low-income populations suggested that
quality care is associated only with known care that is regulated. This was
not our intent. Our discussion represents comments made by CCR&R staff
and state and local officials in response to questions we asked to answer
the second review objective: to identify other challenges, besides possible
gaps between supply and demand, that low-income families face in
accessing child care. The challenges mentioned most often by these
officials included issues about child care quality and affordability, as well
as the availability of nonstandard-hour care and transportation, all of
which are discussed in the latter part of the report.

HHS,the states, and CCR&Rs also provided technical comments, which we
addressed in the report, as appropriate.


As agreed with your office, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 7 days from the
date of this letter. At that time, we will send copies of this report to the
Secretary of Health and Human Services; the Chairmen and Ranking
Minority Members of the House Committees on Ways and Means and
Education and the Workforce; and the Chairmen and Ranking Minority
Members of the Senate Committees on Finance and Labor and Human
Resources. We will also make copies available to others on request.




Page 19                       GAO/HEHS-97-75 Welfare Reform and Child Care Supply
B-270237




If you or your staff have any questions about this report, please contact me
on (202) 512-7125. Other staff who contributed to this report are listed in
appendix IV.

Sincerely yours,




Mark V. Nadel
Associate Director
Income Security Issues




Page 20                      GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Page 21   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Contents



Letter                                                                                             1


Appendix I                                                                                        26
                       Supply Data                                                                26
Scope and              Demand Data                                                                28
Methodology            Estimating Percentage of Demand That Could Be Met by Supply                30
                         of Known Child Care
                       Estimating New Demand Under Welfare Work Requirements                      30
                       Estimating Poor and Nonpoor Supply of Known Care                           31
                       Estimating Poor and Nonpoor Demand for Care                                32

Appendix II                                                                                       34
                       Baltimore City, Maryland                                                   34
Child Care at Four     Chicago, Illinois                                                          36
Sites We Reviewed      Benton County, Oregon                                                      38
                       Linn County, Oregon                                                        40

Appendix III                                                                                      44

Comments From the
Department of Health
and Human Services
Appendix IV                                                                                       47

GAO Contacts and
Staff
Acknowledgments
Related GAO Products                                                                              48


Tables                 Table 1: Estimates of the Percentage of Child Care Demand That              9
                         Could Be Met by Currently Known Supply for Various Age
                         Groups, 1996-2002
                       Table 2: Estimated Gaps Between Known Supply and Demand for                10
                         Child Care for Various Age Groups, 1996-2002
                       Table 3: Estimates of the Percentage of Child Care Demand That             11
                         Could Be Met by Currently Known Supply for Poor and Nonpoor
                         Areas, 1996-2002




                       Page 22                    GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Contents




Table 4: Median Weekly Price for Known Child Care in Poor and               13
  Nonpoor Areas
Table II.1: Total Known Providers by Type of Setting in Poor and            34
  Nonpoor Areas, Baltimore City
Table II.2: Total Known Providers by Age Group in Poor and                  34
  Nonpoor Areas, Baltimore City
Table II.3: Percentage of Known Providers That Offer Care                   34
  During Nonstandard Hours in Poor and Nonpoor Areas,
  Baltimore City
Table II.4: Percentage of Known Providers That Currently                    35
  Provide Special Needs Care in Poor and Nonpoor Areas,
  Baltimore City
Table II.5: Total Known Provider Spaces by Type of Setting in               35
  Poor and Nonpoor Areas, Baltimore City
Table II.6: Known Spaces for Infants by Type of Setting in Poor             35
  and Nonpoor Areas, Baltimore City
Table II.7: Known Spaces for Preschool Children by Type of                  35
  Setting in Poor and Nonpoor Areas, Baltimore City
Table II.8: Known Spaces for School-Aged Children by Type of                36
  Setting in Poor and Nonpoor Areas, Baltimore City
Table II.9: Total Known Providers by Type of Setting in Poor and            36
  Nonpoor Areas, Chicago
Table II.10: Total Known Providers by Age Group in Poor and                 37
  Nonpoor Areas, Chicago
Table II.11: Percentage of Known Providers That Offer Care                  37
  During Nonstandard Hours in Poor and Nonpoor Areas, Chicago
Table II.12: Percentage of Known Providers That Have Had                    37
  Experience Caring for Children With Special Needs in Poor and
  Nonpoor Areas, Chicago
Table II.13: Total Known Provider Spaces by Type of Setting in              37
  Poor and Nonpoor Areas, Chicago
Table II.14: Known Spaces for Infants by Type of Setting in Poor            38
  and Nonpoor Areas, Chicago
Table II.15: Known Spaces for Preschool Children by Type of                 38
  Setting in Poor and Nonpoor Areas, Chicago
Table II.16: Known Spaces for School-Aged Children by Type of               38
  Setting in Poor and Nonpoor Areas, Chicago
Table II.17: Total Known Providers by Type of Setting in Poor and           39
  Nonpoor Areas, Benton County
Table II.18: Total Known Providers by Age Group in Poor and                 39
  Nonpoor Areas, Benton County




Page 23                     GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Contents




Table II.19: Percentage of Known Providers That Offer Care                   39
  During Nonstandard Hours in Poor and Nonpoor Areas, Benton
  County
Table II.20: Total Known Provider Spaces by Type of Setting in               39
  Poor and Nonpoor Areas, Benton County
Table II.21: Known Spaces for Infants by Type of Setting in Poor             40
  and Nonpoor Areas, Benton County
Table II.22: Known Spaces for Preschool Children by Type of                  40
  Setting in Poor and Nonpoor Areas, Benton County
Table II.23: Known Spaces for School-Aged Children by Type of                40
  Setting in Poor and Nonpoor Areas, Benton County
Table II.24: Total Known Providers by Type of Setting in Poor and            41
  Nonpoor Areas, Linn County
Table II.25: Total Known Providers by Age Group in Poor and                  41
  Nonpoor Areas, Linn County
Table II.26: Percentage of Known Providers That Offer Care                   41
  During Nonstandard Hours in Poor and Nonpoor Areas, Linn
  County
Table II.27: Total Known Provider Spaces by Type of Setting in               42
  Poor and Nonpoor Areas, Linn County
Table II.28: Known Spaces for Infants by Type of Setting in Poor             42
  and Nonpoor Areas, Linn County
Table II.29: Known Spaces for Preschool Children by Type of                  42
  Setting in Poor and Nonpoor Areas, Linn County
Table II.30: Known Spaces for School-Aged Children by Type of                43
  Setting in Poor and Nonpoor Areas, Linn County




Abbreviations

AFDC       Aid to Families With Dependent Children
BLS        Bureau of Labor Statistics
CCDBG      Child Care Development Block Grant
CCR&R      child care resource and referral agencies
CPS        Current Population Survey
FSA        Family Support Act
HHS        Department of Health and Human Services
JOBS       Job Opportunities and Basic Skills Training
NCCS       National Child Care Survey
SIPP       Survey on Income and Program Participation


Page 24                      GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Page 25   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Appendix I

Scope and Methodology


              This appendix provides more detail about the methods we used to arrive
              at our estimates of the supply of and demand for child care at our four
              sites. Our measurement of the gap between the current known supply and
              current, as well as future, demand for family child care homes and centers
              is based on a static model of an inherently dynamic process. As the
              demand for child care increases, economic theory would predict that over
              time the supply of care will increase as well, until the two are once again
              in equilibrium.25 Our model provides a snapshot of a point in time at which
              demand has increased but supply has not yet moved to meet it. Thus, we
              are able to identify, for these four sites, the amount of specific types of
              known child care and where it is located, and to predict where care will be
              needed in the future.

              We performed our work between April and December 1996 in accordance
              with generally accepted government auditing standards. We did not
              independently verify the child care supply data provided by the child care
              resource and referral agencies (CCR&R).


              The starting point for our work was the databases provided by the three
Supply Data   CCR&Rs for the four sites reviewed: Baltimore City, Maryland; Chicago,
              Illinois; and Benton and Linn counties, Oregon.26 We used site-specific
              supply data instead of nationwide data because local supply data were
              more readily available, current, and comprehensive, thus improving the
              accuracy of our estimates. Additionally, site-specific supply data were
              needed to be able to examine supply differences between poor and
              nonpoor areas.27


              25
                Economic theory also would predict that if the child care supply curve was upward sloping, the price
              of care would rise as the demand for care increased. This would make measuring the gap between
              supply and demand even more difficult, since we would also have to take into account changing
              prices. Our model of the child care market assumes that the long-run supply curve for child care is
              essentially horizontal; that is, the price of care does not increase with an increase in demand. While
              this assumption might be unusual for more standard commodities, it is a fair description of the
              behavior of the child care market in recent years. While the demand for child care has increased
              dramatically, the price of care, adjusted for inflation, has remained approximately the same.
              26
               The Child Care Bureau at the Administration for Children and Families, U.S. Department of Health
              and Human Services (HHS), and the National Association of Child Care Resource and Referral
              Agencies helped identify a potential list of sites.
              27
                Child care supply data for Baltimore City were provided by the Maryland Committee for Children,
              Inc., a private, nonprofit, community organization. Its database contains information on regulated child
              care and early education programs throughout Maryland, the District of Columbia, and Northern
              Virginia. Data for Benton and Linn counties, Oregon, were provided by Family Connections, a CCR&R
              located at Linn and Benton County Community College. Chicago data were provided by the Day Care
              Action Council, Chicago, Illinois. All three CCR&Rs are members of the Child Care Research
              Partnerships funded by the Child Care Bureau in HHS.



              Page 26                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                              Appendix I
                              Scope and Methodology




Identifying and Classifying   The CCR&Rs’ databases included a wealth of information about currently
Providers                     active known child care providers in their service delivery areas, including
                              their hours of operation, the ages of children that they care for, and the
                              fees that they charge. We first designated each child care provider in each
                              database as either a center or a family day care provider. From the original
                              classifications provided by the CCR&Rs, our definition of center care
                              included a broad and diverse range of full- and part-time programs, such as
                              Head Start and Maryland’s Extended Elementary Education Program; our
                              definition of family day care included both family day care and group
                              homes.28 Providers that did not fall into one of these categories, such as
                              summer camps and providers of care during vacation time only, were
                              excluded from our analysis. In addition, we excluded parental care,
                              relative care, and care provided in the child’s own home.

                              We further identified each provider in each database by the ages and
                              number of children it was willing or licensed to care for. We grouped
                              children into three age categories: infants, birth to 23 months; preschool,
                              aged 24 to 71 months; and school-aged, aged 6 through 12 years. We also
                              identified those providers that were qualified to care for children with
                              special needs and those who were willing to care for children during
                              nonstandard work hours, such as on the weekend, or before 6 a.m. and
                              after 6:30 p.m.


Determining Child Care        The CCR&R databases varied, across sites and types of providers, with
Capacity                      respect to the information they contained on the capacity of each child
                              care provider.29 Capacity data by each age group were available at all four
                              sites for center-based care. However, such data for family child care were
                              not always available from each site. In Chicago, capacity data for each age
                              group were available for family child care. However, in Baltimore City,
                              only data on total capacity and infant capacity were available separately
                              for family child care. After subtracting the infant capacity from the total
                              capacity, we had to estimate preschool and school-aged capacities from
                              the remainder using a formula we developed. This formula was based on
                              the assumption that the ratio of preschool to school-aged children varied
                              slightly depending on the number of infants who were receiving care in the


                              28
                               Group homes were defined as family day care homes that are licensed to care for between 10 and 12
                              children.
                              29
                                At all four sites, center capacity is the maximum number of children allowed at the center by the
                              state. This same definition of capacity is used for family child care homes for Chicago and Baltimore
                              City. For family child care homes in Benton and Linn counties, capacity is the number of children the
                              provider is willing to accept, within the legal limit.



                              Page 27                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                         Appendix I
                         Scope and Methodology




                         home.30 In Benton and Linn counties in Oregon, data for home-based care
                         were available only on total capacity. Because we didn’t have any
                         information specifically on infant capacity for each provider in these
                         counties, we assumed that every provider that accepted infants had a
                         maximum infant capacity of 1. We then subtracted this estimated infant
                         capacity from the total capacity and applied the same formulas used on
                         the Baltimore City data to the remainder to estimate preschool and
                         school-aged capacities.


                         To estimate total demand for family child care homes and centers at our
Demand Data              four sites, we needed to know the number of children through age 12, the
                         percentage of those children with working mothers, and the percentage of
                         children who used either center care or family day care. We therefore used
                         data from a number of different sources, including the 1990 U.S. Census, a
                         1994 update to the Census, the 1995 Current Population Survey (CPS), as
                         well as two surveys of child care usage—the Survey on Income and
                         Program Participation (SIPP) and the National Child Care Survey (NCCS).
                         Because these surveys do not identify whether parents are using care that
                         is known or unknown to the states, we had to estimate total demand for
                         center and family day care.


Estimating Number of     Our estimate of demand for care starts with data from the 1994 update to
Children Aged 12 and     the U.S. Census on the number of children aged 12 and under living at
Under                    each of our four sites, reported by single year of age. We also collected
                         age-specific data from the local welfare offices at each of our four sites on
                         the number of children aged 12 and under who were on the welfare rolls in
                         1995 or 1996. We then subtracted these welfare numbers from the 1994
                         population numbers to create an estimate of the number of nonwelfare
                         children by single year of age in each site.


Estimating Number of     The 1995 CPS provided the percentage of children, at each age, with
Children With Working    working and nonworking mothers in each state. We then applied these
Mothers and Nonworking   state-specific percentages to the number of nonwelfare children
                         determined above, to project numbers of nonwelfare children, at each age,
Mothers                  with working and nonworking mothers at each site. Because of
                         methodological constraints, the children on the welfare rolls were all

                         30
                           We developed the following formula on the basis of discussions with CCR&R officials about the
                         average capacity by age among their provider populations. If infant capacity was 2, the preschool to
                         school-aged ratio was .66/.33; if infant capacity was 1, the ratio was .71/.29; if infant capacity was 0, the
                         ratio was .75/.25.



                         Page 28                                    GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                           Appendix I
                           Scope and Methodology




                           assumed to have nonworking mothers, although we recognize that some
                           mothers on welfare also work. We added the number of children on
                           welfare to the estimate of children of nonwelfare, nonworking mothers to
                           arrive at an estimate of total number of children with nonworking
                           mothers.


Estimating Percentage of   We based our estimates of the percentage of children at each age using
Children Using Centers     center care or family day care, respectively, on two different national
and Family Day Care        surveys. We used the 1993 SIPP data to determine the percentage of
                           children from birth to age 4 using center care and the percentage using
                           family day care. Because the SIPP data had been collected only for working
                           mothers, we had to estimate a usage rate for the children of nonworking
                           mothers as well. To do this, we used 1990 NCCS data, which included both
                           working and nonworking mothers in the sample, to create a ratio of the
                           child care usage rates of children of nonworking mothers to those of
                           working mothers, at each age from birth through 12, for center and family
                           day care separately. We multiplied these ratios by the usage rates for
                           children of working mothers from the SIPP to impute usage rates for
                           children of nonworking mothers in 1993.

                           Because of a design flaw in the questionnaire, the 1993 SIPP data seriously
                           underestimate child care usage rates of school-aged children.31 Therefore,
                           we used the 1990 NCCS data to determine the percentage of children aged 5
                           through 12, with working and nonworking mothers, who used center care
                           and family day care. To compensate for the difference in child care use
                           patterns between 1989-90 and 1993, we applied an adjustment factor to the
                           NCCS data equal to the percentage change in the estimates of center care
                           and family day care use between 1988 and 1993 SIPP data to inflate these
                           1990 figures to what they would have been in 1993.32


Estimating Numbers of      We then multiplied our estimates of the percentages of children using each
Children Using Centers     type of child care, at each age, by the estimated number of children at each
and Family Day Care        age at each site to yield an estimate of the number of children, at each age,



                           31
                            Census Bureau analysts reported that the SIPP child care questionnaire has since been revised to
                           better capture the child care usage of school-aged children.
                           32
                             Before 1993, the SIPP data were collected in 1988 and then again in 1991. The NCCS data were
                           collected in 1989-90. We determined that it was better to use the 1988 SIPP and possibly overestimate
                           the change in usage rates, rather than use the 1991 data and run the risk of underestimating the
                           change.



                           Page 29                                 GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                        Appendix I
                        Scope and Methodology




                        expected to be in each type of care. This is our estimated demand for child
                        care.33


                        Having developed estimates of the supply of known family child care
Estimating Percentage   homes and centers at our four sites and the demand for these types of
of Demand That          care, our final step was to compare, for each age category, the estimated
Could Be Met by         number of child care spaces available and the estimated number of
                        children requiring each type of care for each site. The difference between
Supply of Known         these two estimates is defined as the gap between the current supply of
Child Care              known family child care homes and centers and the total demand for these
                        types of care at each site. The ratio of the number of spaces available to
                        the number of spaces demanded is defined as the estimated percentage of
                        demand for center and family child care that could be met by the
                        estimated current supply of known care. We calculated this percentage for
                        child care overall, as well as for age-specific child care, at each site.


                        To project the possible new demand created as a result of welfare reform,
Estimating New          we assumed that the percentage of children currently on the welfare rolls
Demand Under            who would need child care under welfare reform would be equal to the
Welfare Work            percentage of mothers moving from welfare to the workforce. At every
                        age, we estimated that 25 percent, 35 percent, or 50 percent of the children
Requirements            would need child care because their mothers were now required to work
                        or participate in work-related activities. These percentages are based on
                        the work requirements of the Personal Responsibility and Work
                        Opportunity Reconciliation Act of 1996 for 1997, 1999, and 2002 .

                        We then compared these estimates of the demand for child care after the
                        welfare reform work requirements go into effect with the previously
                        estimated supply of known child care at each site. We made these
                        comparisons for age-specific categories of child care as well as for child
                        care as a whole. These comparisons did not take into account the natural
                        increase in the supply of care that an increase in demand would eventually
                        engender. Our estimate of how capable the current supply of known child
                        care would be of meeting the expected increased demand was intended to
                        illustrate how much the supply of known child care would have to grow to
                        maintain or even increase the percentage of demand that it currently is
                        capable of meeting.

                        33
                         Because of data limitations, our calculations underestimate the number of nonwelfare mothers who
                        are working and overestimate the number who are not working. At the same time, our calculations
                        overestimate the number of welfare mothers who are not working by assuming that all welfare
                        mothers do not work.



                        Page 30                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                            Appendix I
                            Scope and Methodology




                            Because the increase in demand for child care resulting from welfare
Estimating Poor and         reform is expected to come primarily from poor parents, the location of
Nonpoor Supply of           the current supply of child care is a relevant issue. While child care
Known Care                  markets are not completely segregated so that poor parents purchase care
                            only within poor areas and nonpoor parents within nonpoor areas, most
                            parents prefer to use providers who are close to home. Further analysis of
                            the total child care demand and supply of known child care at our four
                            sites involved separating each site into poor and nonpoor areas.


Determining Census Tracts   To categorize providers as being located in either poor or nonpoor areas at
                            each site, we needed to know the census tract number for the location or
                            residence of each provider. Baltimore City’s CCR&R included the census
                            tract number for each provider in its database. The databases for both the
                            Chicago and Oregon CCR&Rs, however, included only the name of the
                            school nearest the location or residence of each provider. For Chicago and
                            the two Oregon counties, we therefore used the Tiger/Census Tract Street
                            Index34 to determine the census tract number for the school closest to the
                            provider and assigned that census tract to the provider. In cases in which
                            Chicago providers had not given accurate school information, the CCR&R
                            released the actual addresses of the providers to us. With this information,
                            we were able to determine and assign to each a census tract number.


Designating Poor and        We analyzed the census tracts within each site by their median household
Nonpoor Census Tracts       income. In some cases, we also looked at the percentage of households on
                            welfare in each tract.35 Each census tract was then designated as either
                            poor or nonpoor. The criteria for this designation varied somewhat by site,
                            especially with respect to the percentage of households on welfare in each
                            tract. In general, those census tracts with a median household income at
                            or below $27,750 (slightly over 200 percent of the federal poverty level for
                            a family of four in 1995) were defined as being in poor areas, and those
                            with median household income above $27,750 were defined as being in
                            nonpoor areas. This criterion was sometimes overridden, however, when
                            examined in conjunction with the criterion on percentage of households
                            on welfare.




                            34
                             The Tiger/Census Tract Street Index is a CD-ROM database that matches addresses with census tract
                            numbers.
                            35
                              These data came from the 1990 U.S Census.



                            Page 31                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                            Appendix I
                            Scope and Methodology




Estimating Poor and
Nonpoor Demand for
Care

Determining Percentage of   We used 1990 Census data by census tract for each of the four sites to
Children in Poor and        estimate the percentages of children from birth through age 12 who were
Nonpoor Areas               living in poor and nonpoor areas in each site.

Determining Percentage of   The 1995 CPS provided the percentage of poor and nonpoor children with
Children With Working and   working and nonworking mothers in each state. We then applied these
Nonworking Mothers in       state-specific percentages to the numbers of poor and nonpoor nonwelfare
                            children for each site to project numbers of nonwelfare children with
Poor and Nonpoor Areas      working and nonworking mothers separately for poor and nonpoor areas
                            of each site. We then included the number of children on the welfare rolls
                            in the estimate of children in poor areas with nonworking mothers.


No Difference in Child      Because of data limitations, we estimated that for each child care setting
Care Usage Rates Assumed    the usage rates of poor children are the same as those of nonpoor
for Poor and Nonpoor        children. This is a shortcoming of our methodology, however, because
                            poor and nonpoor parents use different types of child care at different
Areas                       rates. In particular, lower-income parents are more likely to use relative
                            care, which is either free or much lower in price than market care (center
                            care and family day care). However, one outcome of welfare reform may
                            be a decrease in the availability of free relative care for this population,
                            since more people, including perhaps those caregiving relatives, will be
                            required to work. In addition, while in the past the increase in demand for
                            child care has not affected its long-term price, another short-term effect of
                            the increase in demand for care attributable to welfare reform may be to
                            drive up the price of care for all parents, but for low-income parents in
                            particular. Both of these possibilities may result in poor parents using
                            more market care than they have in the past.


Estimating Numbers of       We then multiplied our estimates of the percentages of children using each
Poor and Nonpoor            type of child care at each age by the estimated number of poor and
Children Using Centers      nonpoor children at each age at each site to yield an estimate of the
                            number of poor and nonpoor children, at each age, expected to demand
and Family Day Care         each type of care.




                            Page 32                       GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                             Appendix I
                             Scope and Methodology




Estimating Percentage of     We compared the estimated number of known child care spaces available
Current and Projected        within each age category and the estimated number of children currently
Demand That Could Be         demanding each type of care in each age category for each site. This was
                             done separately for poor and nonpoor areas. In addition, we estimated the
Met by Current Known         new demand for child care in the poor areas that is likely to be caused by
Supply in Poor and           the move of more mothers from welfare to work using the same rates of
Nonpoor Areas                labor force participation we assumed above.


Calculating Median Weekly    To calculate the median weekly price of child care, we again divided each
Price of Child Care by Age   site-specific database according to the type of provider (family day care or
of Child for Each Site       center care) and the age category of children each provider served (infant,
                             preschool, or school-aged). Where possible, we identified a full-time
                             weekly price of care for each provider in each age/type category. When a
                             full-time weekly price was not available, we estimated it using the given
                             part-time hourly rate and a standard number of hours for full-time care of
                             45 hours for infants and preschoolers and 25 hours for school-aged
                             children. If no price information was available for a specific provider, we
                             dropped it from the sample when calculating the median weekly price.
                             Thus, a median weekly price was calculated for each age category and
                             type of child care setting, for the total number of child care providers that
                             provided price information, and for each site.




                             Page 33                       GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Appendix II

Child Care at Four Sites We Reviewed


                                       Baltimore City has over 700,000 residents and ranks thirteenth among U.S.
Baltimore City,                        cities in population. According to the Bureau of Labor Statistics (BLS),
Maryland                               5.5 percent of Baltimore City’s labor force was unemployed in
                                       September 1996. While 15 percent of the state’s population lives in
                                       Baltimore City, about 47 percent of the state’s Aid to Families With
                                       Dependent Children (AFDC) caseload, or about 88,000 participants, resides
                                       there. Of the city’s AFDC caseload, about 64 percent, or 56,554, is children
                                       12 and under. These children represent 41 percent of the total population
                                       of children 12 and under, about 137,000, living in Baltimore City.36

Table II.1: Total Known Providers by
Type of Setting in Poor and Nonpoor                   Center providers              Family providers               Total providers
Areas, Baltimore City                                 Number Percentage             Number Percentage             Number Percentage
                                       Poor
                                       areas               184              52           654              54           838              54
                                       Nonpoor
                                       areas               171              48           557              46           728              46
                                       Total               355            100          1,211            100          1,566            100
                                       Note: Percentages are rounded to the nearest whole number. Centers include those providers
                                       classified as center care, a group program, or an Extended Elementary Education Program.



Table II.2: Total Known Providers by
Age Group in Poor and Nonpoor                          Infant providers           Preschool providers          School-aged providers
Areas, Baltimore City                                 Number Percentage             Number Percentage             Number Percentage
                                       Poor
                                       areas               552              53           733              53           629              54
                                       Nonpoor
                                       areas               489              47           647              47           529              46
                                       Total             1,041            100          1,380            100          1,158            100
                                       Note: Percentages are rounded to the nearest whole number.



Table II.3: Percentage of Known
Providers That Offer Care During                                                                                Providers in nonpoor
Nonstandard Hours in Poor and                                                    Providers in poor areas               areas
Nonpoor Areas, Baltimore City                                                       Number Percentage             Number Percentage
                                       Nonstandard hours                                 101              12             94             13
                                       Total                                             838            100            728            100
                                       Note: Percentages are rounded to the nearest whole number.



                                       36
                                        The source years for these data and those in the tables that follow ranged from 1994 to 1996;
                                       comparable data for a common point in time were unavailable. These data are presented to provide a
                                       general overview of the economic environment and welfare population of the city.



                                       Page 34                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                          Appendix II
                                          Child Care at Four Sites We Reviewed




Table II.4: Percentage of Known
Providers That Currently Provide                                                                                   Providers in nonpoor
Special Needs Care in Poor and                                                      Providers in poor areas               areas
Nonpoor Areas, Baltimore City                                                          Number Percentage              Number Percentage
                                          Special needs                                      109             13             90             12
                                          Total                                              838            100            728             100
                                          Note: Percentages are rounded to the nearest whole number.



Table II.5: Total Known Provider
Spaces by Type of Setting in Poor and                    Center providers              Family providers                Total providers
Nonpoor Areas, Baltimore City                             Spaces Percentage             Spaces Percentage             Spaces Percentage
                                          Poor
                                          areas             7,779              50          4,009             53        11,788              51
                                          Nonpoor
                                          areas             7,626              50          3,587             47        11,213              49
                                          Total            15,405             100          7,596            100        23,001              100
                                          Note: Percentages are rounded to the nearest whole number. Centers include those providers
                                          classified as center care, a group program, or an Extended Elementary Education Program.
                                          Family spaces in Baltimore City were estimated. See app. I for the details of our methodology.



Table II.6: Known Spaces for Infants by
Type of Setting in Poor and Nonpoor                      Center providers              Family providers                Total providers
Areas, Baltimore City                                     Spaces Percentage             Spaces Percentage             Spaces Percentage
                                          Poor
                                          areas               110              50            935             52          1,045             52
                                          Nonpoor
                                          areas               108              50            849             48            957             48
                                          Total               218             100          1,784            100          2,002             100
                                          Note: Percentages are rounded to the nearest whole number. Centers include those providers
                                          classified as center care, a group program, or an Extended Elementary Education Program.



Table II.7: Known Spaces for
Preschool Children by Type of Setting                    Center providers              Family providers                Total providers
in Poor and Nonpoor Areas, Baltimore                      Spaces Percentage             Spaces Percentage             Spaces Percentage
City
                                          Poor
                                          areas             5,877              53          2,121             53          7,998             53
                                          Nonpoor
                                          areas             5,231              47          1,884             47          7,115             47
                                          Total            11,108             100          4,005            100        15,113              100
                                          Note: Percentages are rounded to the nearest whole number. Centers include those providers
                                          classified as center care, a group program, or an Extended Elementary Education Program.




                                          Page 35                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                       Appendix II
                                       Child Care at Four Sites We Reviewed




Table II.8: Known Spaces for
School-Aged Children by Type of                       Center providers              Family providers               Total providers
Setting in Poor and Nonpoor Areas,                     Spaces Percentage             Spaces Percentage             Spaces Percentage
Baltimore City
                                       Poor
                                       areas             1,792              44           953              53         2,745              47
                                       Nonpoor
                                       areas             2,287              56           854              47         3,141              53
                                       Total             4,079            100          1,807            100          5,886            100
                                       Note: Percentages are rounded to the nearest whole number. Centers include those providers
                                       classified as center care, a group program, or an Extended Elementary Education Program.




                                       Chicago has about 2.8 million residents and ranks third among U.S. cities
Chicago, Illinois                      in population. According to BLS, 4.8 percent of Chicago’s labor force was
                                       unemployed in September 1996. While only 24 percent of the state’s
                                       population lives in Chicago, about 55 percent of the state’s AFDC caseload,
                                       or 351,000 participants, resides there. Of the city’s AFDC caseload, about
                                       60 percent, or 219,489, is children 12 and under. These children represent
                                       about 40 percent of the total population of children, about 561,000, living
                                       in Chicago.37

Table II.9: Total Known Providers by
Type of Setting in Poor and Nonpoor                   Center providers              Family providers               Total providers
Areas, Chicago                                        Number Percentage             Number Percentage             Number Percentage
                                       Poor
                                       areas               432              56           405              39           837              46
                                       Nonpoor
                                       areas               343              44           624              61           967              54
                                       Total               775            100          1,029            100          1,804            100
                                       Note: Percentages are rounded to the nearest whole number.




                                       37
                                        The source years for these data and those in the tables that follow ranged from 1992 to 1996;
                                       comparable data for a common point in time were unavailable. These data are presented to provide a
                                       general overview of the economic environment and welfare population of the city.



                                       Page 36                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                         Appendix II
                                         Child Care at Four Sites We Reviewed




Table II.10: Total Known Providers by
Age Group in Poor and Nonpoor                            Infant providers          Preschool providers        School-aged providers
Areas, Chicago                                         Number Percentage             Number Percentage          Number Percentage
                                         Poor
                                         areas              382              39          801             46         390            41
                                         Nonpoor
                                         areas              608              61          940             54         569            59
                                         Total              990             100        1,741            100         959            100
                                         Note: Percentages are rounded to the nearest whole number.



Table II.11: Percentage of Known
Providers That Offer Care During                                                                              Providers in nonpoor
Nonstandard Hours in Poor and                                                     Providers in poor areas            areas
Nonpoor Areas, Chicago                                                               Number Percentage          Number Percentage
                                         Nonstandard hours                               167             20         225            23
                                         Total                                           837            100         967            100
                                         Note: Percentages are rounded to the nearest whole number.



Table II.12: Percentage of Known
Providers That Have Had Experience                                                                            Providers in nonpoor
Caring for Children With Special Needs                                            Providers in poor areas            areas
in Poor and Nonpoor Areas, Chicago                                                   Number Percentage          Number Percentage
                                         Special needs                                   419             50         371            38
                                         Total                                           837            100         967            100
                                         Note: Percentages are rounded to the nearest whole number.



Table II.13: Total Known Provider
Spaces by Type of Setting in Poor and                  Center providers              Family providers            Total providers
Nonpoor Areas, Chicago                                   Spaces Percentage           Spaces Percentage           Spaces Percentage
                                         Poor
                                         areas            25,529             57        2,645             37      28,174            54
                                         Nonpoor
                                         areas            19,107             43        4,417             63      23,524            46
                                         Total            44,636            100        7,062            100      51,698            100
                                         Note: Percentages are rounded to the nearest whole number.




                                         Page 37                              GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                        Appendix II
                                        Child Care at Four Sites We Reviewed




Table II.14: Known Spaces for Infants
by Type of Setting in Poor and                         Center providers              Family providers               Total providers
Nonpoor Areas, Chicago                                  Spaces Percentage             Spaces Percentage             Spaces Percentage
                                        Poor
                                        areas               218              28           927              38         1,145              35
                                        Nonpoor
                                        areas               562              72         1,519              62         2,081              65
                                        Total               780            100          2,446            100          3,226            100
                                        Note: Percentages are rounded to the nearest whole number.



Table II.15: Known Spaces for
Preschool Children by Type of Setting                  Center providers              Family providers               Total providers
in Poor and Nonpoor Areas, Chicago                      Spaces Percentage             Spaces Percentage             Spaces Percentage
                                        Poor
                                        areas            20,294              55         1,537              38        21,831              54
                                        Nonpoor
                                        areas            16,311              45         2,528              62        18,839              46
                                        Total            36,605            100          4,065            100         40,670            100
                                        Note: Percentages are rounded to the nearest whole number.



Table II.16: Known Spaces for
School-Aged Children by Type of                        Center providers              Family providers               Total providers
Setting in Poor and Nonpoor Areas,                      Spaces Percentage             Spaces Percentage             Spaces Percentage
Chicago
                                        Poor
                                        areas             5,017              69           181              33         5,198              67
                                        Nonpoor
                                        areas             2,234              31           370              67         2,604              33
                                        Total             7,251            100            551            100          7,802            100
                                        Note: Percentages are rounded to the nearest whole number.




                                        Benton County has about 75,500 residents or about 2 percent of the state’s
Benton County,                          population. In September 1996, 2.4 percent of Benton County’s labor force
Oregon                                  was unemployed. About 3 percent, or 3,153, of the state’s AFDC population
                                        resides in Benton County. Of the county’s AFDC caseload, about 46 percent
                                        is children 12 and under. These 1,446 children represent about 12 percent
                                        of the total population, about 11,909, of children living in Benton County.38


                                        38
                                         The source years for these data and those in the tables that follow ranged from 1994 to 1996;
                                        comparable data for a common point in time were unavailable. These data are presented to provide a
                                        general overview of the economic environment and welfare population of the county.



                                        Page 38                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                        Appendix II
                                        Child Care at Four Sites We Reviewed




Table II.17: Total Known Providers by
Type of Setting in Poor and Nonpoor                    Center providers             Family providers              Total providers
Areas, Benton County                                   Number Percentage            Number Percentage            Number Percentage
                                        Poor
                                        areas                10             32            54             36            64             35
                                        Nonpoor
                                        areas                21             68            96             64           117             65
                                        Total                31           100            150           100            181             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.18: Total Known Providers by
Age Group in Poor and Nonpoor                          Infant providers           Preschool providers         School-aged providers
Areas, Benton County                                   Number Percentage            Number Percentage            Number Percentage
                                        Poor
                                        areas                46             35            60             37            48             33
                                        Nonpoor
                                        areas                84             65           104             63            99             67
                                        Total               130           100            164           100            147             100
                                        Note: Percentages are rounded to the nearest whole number.



Table II.19: Percentage of Known
Providers That Offer Care During                                                                              Providers in nonpoor
Nonstandard Hours in Poor and                                                    Providers in poor areas             areas
Nonpoor Areas, Benton County                                                        Number Percentage            Number Percentage
                                        Nonstandard hour                                  12             19            15             13
                                        Total                                             64           100            117             100
                                        Note: Percentages are rounded to the nearest whole number.



Table II.20: Total Known Provider
Spaces by Type of Setting in Poor and                  Center providers             Family providers              Total providers
Nonpoor Areas, Benton County                           Spaces Percentage            Spaces Percentage            Spaces Percentage
                                        Poor
                                        areas               458             33           301             38           759             35
                                        Nonpoor
                                        areas               918             67           499             62         1,417             65
                                        Total             1,376           100            800           100          2,176             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.




                                        Page 39                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                        Appendix II
                                        Child Care at Four Sites We Reviewed




Table II.21: Known Spaces for Infants
by Type of Setting in Poor and                         Center providers             Family providers              Total providers
Nonpoor Areas, Benton County                           Spaces Percentage            Spaces Percentage            Spaces Percentage
                                        Poor
                                        areas                80             44            42             34           122             40
                                        Nonpoor
                                        areas               100             56            81             66           181             60
                                        Total               180           100            123           100            303             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.22: Known Spaces for
Preschool Children by Type of Setting                  Center providers             Family providers              Total providers
in Poor and Nonpoor Areas, Benton                      Spaces Percentage            Spaces Percentage            Spaces Percentage
County
                                        Poor
                                        areas               207             34           227             39           434             36
                                        Nonpoor
                                        areas               403             66           357             61           760             64
                                        Total               610           100            584           100          1,194             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.23: Known Spaces for
School-Aged Children by Type of                        Center providers             Family providers              Total providers
Setting in Poor and Nonpoor Areas,                     Spaces Percentage            Spaces Percentage            Spaces Percentage
Benton County
                                        Poor
                                        areas               171             29            32             34           203             30
                                        Nonpoor
                                        areas               415             71            62             66           477             70
                                        Total               586           100             94           100            680             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.




                                        Linn County has more than 98,000 residents. As of September 1996,
Linn County, Oregon                     5.5 percent of Linn County’s labor force was unemployed. While only
                                        3 percent of the state’s population lives in Linn County, about 7 percent of
                                        state’s AFDC caseload, about 7,800, resides there. Of the county’s recipients,
                                        almost half are children 12 and under. These children represent about




                                        Page 40                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                        Appendix II
                                        Child Care at Four Sites We Reviewed




                                        20 percent of the total population of children, 18,417, living in Linn
                                        County.39

Table II.24: Total Known Providers by
Type of Setting in Poor and Nonpoor                    Center providers              Family providers               Total providers
Areas, Linn County                                     Number Percentage             Number Percentage             Number Percentage
                                        Poor
                                        areas                  5             25           114              43           119              42
                                        Nonpoor
                                        areas                 15             75           150              57           165              58
                                        Total                 20           100            264            100            284            100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.25: Total Known Providers by
Age Group in Poor and Nonpoor                           Infant providers           Preschool providers          School-aged providers
Areas, Linn County                                     Number Percentage             Number Percentage             Number Percentage
                                        Poor
                                        areas                 94             42           113              42           112              44
                                        Nonpoor
                                        areas               131              58           158              58           140              56
                                        Total               225            100            271            100            252            100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.26: Percentage of Known
Providers That Offer Care During                                                                                 Providers in nonpoor
Nonstandard Hours in Poor and                                                     Providers in poor areas               areas
Nonpoor Areas, Linn County                                                           Number Percentage             Number Percentage
                                        Nonstandard hour                                    49             41             53             32
                                        Total                                             119            100            165            100
                                        Note: Percentages are rounded to the nearest whole number.




                                        39
                                         The source years for these data and those in the tables that follow ranged from 1994 to 1996;
                                        comparable data for a common point in time were unavailable. These data are presented to provide a
                                        general overview of the economic environment and welfare population of the county.



                                        Page 41                                GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                        Appendix II
                                        Child Care at Four Sites We Reviewed




Table II.27: Total Known Provider
Spaces by Type of Setting in Poor and                  Center providers             Family providers              Total providers
Nonpoor Areas, Linn County                             Spaces Percentage            Spaces Percentage            Spaces Percentage
                                        Poor
                                        areas               235             27           638             44           873             37
                                        Nonpoor
                                        areas               649             73           822             56         1,471             63
                                        Total               884           100          1,460           100          2,344             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.28: Known Spaces for Infants
by Type of Setting in Poor and                         Center providers             Family providers              Total providers
Nonpoor Areas, Linn County                             Spaces Percentage            Spaces Percentage            Spaces Percentage
                                        Poor
                                        areas                20             25            92             42           112             37
                                        Nonpoor
                                        areas                60             75           129             58           189             63
                                        Total                80           100            221           100            301             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.



Table II.29: Known Spaces for
Preschool Children by Type of Setting                  Center providers             Family providers              Total providers
in Poor and Nonpoor Areas, Linn                        Spaces Percentage            Spaces Percentage            Spaces Percentage
County
                                        Poor
                                        areas                71             19           438             44           509             37
                                        Nonpoor
                                        areas               298             81           553             56           851             63
                                        Total               369           100            991           100          1,360             100
                                        Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                        homes.




                                        Page 42                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                                     Appendix II
                                     Child Care at Four Sites We Reviewed




Table II.30: Known Spaces for
School-Aged Children by Type of                     Center providers             Family providers              Total providers
Setting in Poor and Nonpoor Areas,                  Spaces Percentage            Spaces Percentage            Spaces Percentage
Linn County
                                     Poor
                                     areas               144             33           108             44           252             37
                                     Nonpoor
                                     areas               291             67           140             56           431             63
                                     Total               435           100            248           100            683             100
                                     Note: Percentages are rounded to the nearest whole number. Family child care includes group
                                     homes.




                                     Page 43                               GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Appendix III

Comments From the Department of Health
and Human Services




               Page 44   GAO/HEHS-97-75 Welfare Reform and Child Care Supply
                    Appendix III
                    Comments From the Department of Health
                    and Human Services




Now on pp. 8-10.

Now on pp. 10-12.




Now on pp. 12-14.




                    Page 45                         GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Appendix III
Comments From the Department of Health
and Human Services




Page 46                         GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Appendix IV

GAO Contacts and Staff Acknowledgments


                  David P. Bixler, Assistant Director, (202) 512-7201
GAO Contacts      Janet L. Mascia, Evaluator-in-Charge, (202) 512-7263


                  In addition to those named above, the following individuals made
Acknowledgments   important contributions to this report: Alicia Puente Cackley, Suzanne
                  Sterling, and Rodina Tungol assisted in designing the job, conducting
                  interviews, analyzing data, and writing the report; James Wright and Joel
                  Grossman provided design support; Joan Vogel and Bob DeRoy conducted
                  computer programming; and Steve Machlin provided statistical advice
                  regarding our data analysis.




                  Page 47                      GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Related GAO Products


              Early Childhood Programs: Multiple Programs and Overlapping Target
              Groups (GAO/HEHS-95-4FS, Oct. 31, 1995).

              Welfare to Work: Child Care Assistance Limited; Welfare Reform May
              Expand Needs (GAO/HEHS-95-220, Sept. 21, 1995).

              Early Childhood Programs: Many Poor Children and Strained Resources
              Challenge Head Start (GAO/HEHS-94-169BR, May 17, 1995).

              Early Childhood Centers: Services to Prepare Children for School Often
              Limited (GAO/HEHS-95-21, Mar. 21, 1995).

              Child Care: Child Care Subsidies Increase Likelihood That Low-Income
              Mothers Will Work (GAO/HEHS-95-20, Dec. 30, 1994).

              Child Care: Promoting Quality in Family Child Care (GAO/HEHS-95-93, Dec. 9,
              1994).

              Child Care: Working Poor and Welfare Recipients Face Service Gaps
              (GAO/HEHS-94-87, May 13, 1994).

              Infants and Toddlers: Dramatic Increases in Numbers Living in Poverty
              (GAO/HEHS-94-74, Apr. 7, 1994).

              School-age Demographics: Recent Trends Pose New Educational
              Challenges (GAO/HRD-93-105BR, Aug. 5, 1993).

              Poor Preschool-aged Children: Numbers Increase but Most Not in
              Preschool (GAO/HRD-93-111BR, July 21, 1993).

              Child Care: States Face Difficulties Enforcing Standards and Promoting
              Quality (GAO/HRD-93-13, Nov. 20, 1992).

              Early Childhood Education: What Are the Costs of High-Quality Programs?
              (GAO/HRD-90-43BR, Jan. 24, 1990).




(106708)      Page 48                      GAO/HEHS-97-75 Welfare Reform and Child Care Supply
Ordering Information

The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent of Documents, when
necessary. VISA and MasterCard credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 6015
Gaithersburg, MD 20884-6015

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (301) 258-4066, or TDD (301) 413-0006.

Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu will provide information on
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with "info" in the body to:

info@www.gao.gov

or visit GAO’s World Wide Web Home Page at:

http://www.gao.gov




PRINTED ON    RECYCLED PAPER
United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                 Permit No. G100
Official Business
Penalty for Private Use $300

Address Correction Requested