1Q ’59.48 R46 £000b LGSL UCB Children fliers Balance Work ad Child care under Welfare Reform I» Ichnical Supplement Iwing Up in Poverty Project The recent report, Remember the Children, provides an initial glimpse into how mothers and children are faring under welfare reform. The present supplement puts forward additional technical information. The first section addresses issues related to the number ofchildren moving into nonmaternal forms of child care. Second, we provide additional details on the quality ofcare selected by mothers. Third, we speak to newness of the welfare—to—work push on women with preschool—age children. Finally, we make a few technical corrections to the original report. How Many Children Are Moving into New Child Care Settings? Two major pressures are encouraging single mothers to move into the paid labor force. First, the past decade of steady economic expansion has boosted demand for semi— skilled workers. Second, welfare reforms advanced in the states. from 1993 forward. required that mothers engage real iobs or work preparation activities. These two forces. combined. have greatly reduced the welfare rolls. for their young children, this leads to the question of how many prewhooLage youngsters have been moved into new child—care settings as a direct result ofwelfare reform. ly’emwu/Iei' the Children conservatively estimated this number at 1 million. Our method appears in Figure 51. Column 3 reports on the declining number of children whose parents received cash assistance under the old .v\l5l )( f or new TANF program. In 1994, 9.8 children were benefiting from cash assistance, falling to 6.3 million in 1998, Numbers for the 1999 fiscal year were not available when we made our estimate. Employment rates for women on welfare were creeping up slowly in the pre-1996 period, then jumped to 55925 in 1998 as federal work requirements kicked-in across the states. Simply looking at 1998 figures — yeti/Hg aria/e earlier effects on maternal employment and child care changes ~ we see that 800,000 preschool—age children had mothers who were now engaged in work activities (column 4). In addition. we know that over halfof all women leaving the welfare rolls were working during their first year after exit, resulting in another 1 million children who presumably moved to new child care settings (column 6). This results in an estimate of 1.8 million preschool—age children in 1998 whose mothers faced new work pressure, either by remaining in or exiting welfare programs, and therefore had to find child care settings for the first time or were moved to new providers. This estimate may be too low, as some critics suggest, since it‘s based on annual declines in caseloads and exit from the TANF system. At the satne time successive cohorts of young children are entering, or more importantly not entering, more restrictive welfare systems found in many states. But the movement of cohorts can only be tracked with longitudinal data bases; these administrative records can not support this more complete analysis. Our estimation also may be low in that it fails to take into account further declines in the rolls experienced in fiscal year 1999 and the current fiscal year, 1999—2000. The further ratcheting—up of the share of mothers who must be engaged in work activities will likely move more children into new child care settings, although this effect is moder- ated by the fall in caseloads and declining pressure on the states to put more parents to work. In addition, the estimate may be on the low side since it does not take into account the projected rise in caseloads, prior to 1996, that was witnessed during the post-1983 economic expansion. Until 1994, welfare rolls continued to steadily grow even under conditions of steady job growth, as detailed by House Ways and Means analyst, Ron Haskins.1 The post—1992 state waiver programs and the 1996 federal act have dissuaded many more parents from entering the welfare system who otherwise would have contributed to this secular growth trend. At the same time, the 1.8 million net estimate could be [7ng7, due to the fact that some of this movement into jobs is driven by rising labor demand, not by welfare reform. However, the \X’hite House Council of Economic Advisors argues that cross—state variability in the magnitude of decline in welfare rolls is increasingly attributable to policy change, rather than to job growth} BERKELEY »- YALE SUPPLEMENT Figure 51 Historical trends: Welfare Children WhO likely face new Child care settings [1] [2] Children [3] Employment (4] Children, age [5] Net decline in [6] Children, age 05, [7] Total children, Fiscal receiving cash rate of mothers 0-5, in aided homes children, age 0-5, with employed age 05, likely facing year assistance receiving cash with workng receiving cash mothers, after leaving new child care (millions)1 assistance2 mothers (millions)3 assistance (millions) welfare (millions)d settings (millionsis 1994 9.8 7.6% 0.4 - - 0.4+6 1995 9.3 8.8% 0.4 0.5 - 0.4+ 1996 8.7 10.0%7 0.4 0.6 - 0.4+ 1997 7.9 11.2% 0.4 0.8 0.5 0.9 1998 6.3 33% 0.8 1.6 1.0 1.8 SOURCES: House Ways and Means Committee, "Green Books,” Washing- ton, D.C. Most data tables available on the Web: www.access.gpo.gov/ congress/meO1html. Annual TANF reports, Administration for Children and Families, Department of Health and Human Services, available on the Web: wwwacfidhhs. S. Brauner and P. Loprest, Where are they now? What states' studies of people who left welfare tell us. (Washington, DC: Urban institute, No. A-32, 1999). NOTES: 1. Until 1996 cash assistance was provided through Aid to Families with Dependent Children (AFDC), then this welfare program was renamed to Temporary Assistance for Needy Families (TANF) 2. includes jobs and "work activities," such as attending a job club, for which the mother would have to Find child care. 3. Counts only preschool-age children in those homes with a working Finally, we know that many women on welfare do use child care even when they are not employed. including enrolling their preschoolers in Head Start and state-funded preschool programs. As these mothers move into jobs. they may not change child care arrangements. Thus the 1.8 million one—time estimate may error on the high side. The Quality of Child Care Our design includes experimental data from Connecticut. where women were randomly assigned into the new Jobs First program or to a control group where they remain under the old AFDC rules. In addition, Remember the Children reported on baseline (Wave 1) data for mothers and children in California and Florida. The data must be interpreted carefully. Remember the Chi/die” relates the sequence of behavior intended by welfare reform. That is. the “theory ofaction“ underlying I] GROWiNC- UP IN POVERTY PROJECT A mother, estimated from column 3. in 1994 JUSt under it: or .iii , iniinn receiving AFDC cash assistance were under 6 years of age in (“Ms llit share stood at 40%. 4. Tracking studies of women leaving cash assistance Ila-(rt Hill on till the post-1996 period, that between 51“ ~ and 11 one and 14 months after exit, depending upen Till lili‘ (Hi 3.» llii'l .\ Loprest (1999). We use the midway point, Fl in: the. estimates. ilr‘ t‘iniiloye l .‘iiilml iiliiiniil 5. Sum of columns 4 and 6. 6. We know that many short-term AFDC ietipiont inn»... ii i i l nit jobs after leaving cash assistance But these iiilliilii‘l‘ 'iilil lll ii i . comparable to post-1996 policy conditions 7. Work participation rates For1996 and 199 7' are i-iiw ii. ii nun. (in: trend line established prior to the 1996 rel M iii’- the welfiu‘e-to-work policy eneourages llliilllkl‘a to lind and hold—down jobs. This. in turn, l'L‘tIiiilc‘s that put-tits lind a stable child care arrangement. lapel iinental e\ idenee. as we discovered in Connecticut. shims lli.ii lliis iliigiaiion is occurring: women racing stronger pitssnii s to \VUI'L more frequently use child care. lhis is not surprising. A second—order causal argument requires more rest-art h. We have not examined whether new it “oi-Ling mothers select child care that is ol‘lou e1 quality than the quality or care provided by mothers who doni node. or those \\ ho have more time to find a stable or higher qualin prox ltlt‘l. In Connecticut. we will have the truly experimental data to test this argument. and we will. in (aliloruia and Florida, once two waves or data are collected \\ e \\'lll lie able to see whether the mothers engagement \\ ith the welfare system and/or her work experient e t ontrihuies to the quality ofcare that she selects. l lthei rest-art h teams also will he ahle to address this issue. U.C. BERKELEY LIBRARIES xlllllljljjlyljljlljljll/ Child care quality was uneven —— although not uniformly poor _ long before the most recent welfare changes.3 Welfare reform is placing new strains on a child care “system” in ways that may further thin—out average quality, especially in low—income communities. The important empirical issue is whether family—level selection processes are conditioned by local organizational conditions. For example, mothers coming off welfare may select less stable or lower quality providers, relative to other parents in the same communities. That is, we would like to know whether they are sorting into the low end of an already uneven provider market. Another issue related to quality care is how it varies across settings. Two—thirds of the participating women, as detailed in Remember the Children, had selected home— based child care. including licensed child care homes and kith or kin members; the remaining third selected centers and preschool programs. Quality. regardless of the child care setting, is multi— dimensional. The graphs and tables in section 8 of the technical report detail these indicators. On some gauges of quality —— especially the education level of the adult caregiver. pursuit of additional training by the caregiver, the avail-ability of play and learning materials, and quality of facilities — the centers exhibited substantially higher levels of quality than home—based settings. Bur on other measures — including the amount of one-to— one language and social interaction that we observed between toddler and adult —— home—based settings looked a bit better. Mothers clearly see home—based settings as more flexible, not exhibiting the organizational rigidities that some women associate with centers, such as closing mid—afternoon, just when their work shift begins. Yet overall, home—based care exhibited the lowest levels of quality. The New Work Push on Mothers The current welfare-to-work push is not entirely new. What’s unprecedented is that now almost all single mothers with preschool—age children are subject to work requirements. Prior to 1996, women with children 3 years-old and younger were exempt from work require— ments. And actual work participation rates for women with 4 and 5 year—olds were very low, except for a few local “workfare” experiments. As late as 1994, just 12% of all adults on AFDC were participating in work activities under JOBS; this share equaled 6% in Florida.4 Corrections Figure 8.13 is a pivotal display. It details how the quality of child care providers selected by the participating mothers compares to the quality of providers observed in earlier national studies. We reprint the figure here with one small correction which does not change the basic pattern of findings. An erratum sheet did correct this in most copies of the original report. A few additional corrections: The 1998 level of monthly cash assistance for a family of3 in Connecticut was $543, not $464 as appeared in Table 3.1 in Remember the Children. The child care quality tool, developed by Figure 8.13 Are welfare children exposed to lower quality child care? ECRS or FDCRS score CT centers CA centers I 1998-99 cup 1997-98 cup comparison CA FCCHS CT FCCHS National FCCHS 1995 C00 comparison 1994 FWI comparison BERKELEY — YALE E Mathematic-a Policy Research lnc.. is called the Child— Caregiver Observation Systetn [LT-COS], as discussed in section 8. Figure 8.4 refers to all mothers paying for child care“ inclttding those making copavments. Endnotes ‘ R. Haskins‘ “\V'ellare in a society of permanent work." Paper presented at a conference olithe Joint Center for Poverty Research‘ \V’ashington l).C., September (livanston: Northwestern University" 1909). 3 Council of~ Economic Advisors‘ “The effects or welfare policy and the economic expansion on welfare caseloads: An update.“ (\V'ashington D.C.: the \\"hite House. 1999). 3‘ Sotne critics ot‘our technical report claim that we have long known that the quality of~ child care is poor in low— income communities. In fact subsidized center—based care has been found to be or‘ higher quality than that found in many middle—class communities on some indicators. Yet we have not known whether welfare parents have been able to find attd enter these higher quality centers. See‘ for example, B. Fuller. S. Raudenbnsh, 1.. \Vei, and S. Holloway. ”Can government raise child care quality? The influence ol‘l‘amily demand. poverty. and policy," lid/{tuninUd/ [Z‘I'KI/I/KIfI-(H/ {11/{//)(7//(:)li‘ll/tlzl'V/‘I 1 3 ( 1003):.‘li 2—78. Just one earlier national study olihomerlmsed child care‘ including observational measures ()l‘tllltllll}. has been available prior to the publication oli ly’t‘mtzw/It'z’ r/tt‘ ( '/t/'///n'1/4 ‘ The JOBS welfare reforms. approved by Congress and the President in WSS, required that yyomen \\ ith I and ‘t year—olds be enrolled itt work related acti\ it ies. l’rey iously. no work requirement was in place lor \‘y'tiliicii until their child entered school. Mandatory elilttllnicliis III \a oily activities‘ such as job clttbs and plat etnent in teal itilts remained low and uneven across the states. nnnl HIIIH'I sally applied with the 1990 rel‘ormst Some start s ntm exempt women from work requirements ttnttl lllk n llilJIiI turns (w or 12 months oi‘age. But the tnotlttt s "t lm l- f counting down months to their titnt limit "li : .lsll .inHIH tance, is ticking yyhile they raist lllt n IiIl n.» it Ft‘iedlander and (i. Bttrtless. [7/11 it a t «Um in / wr/t term [Effects (ff'll'i’X/t‘zt't’—r()~ll'iwl‘ Wing/t, , \ tX- ‘th l- lx'ttsst'll Sage“ 190;). l). 0T\X’il.111tll l'llt’tilndi i i t d fin/my //’l lie/fare Reform (\Kalanta/oo: l'pittln: ln lii'fit 1‘“ 'I For more information on the Growing Up in Poverty Project Graduate School ot‘liducation — PACE l'niversity ot‘Calit‘ornia, Berkeley ‘Mfilli S 1 (NH 27323 http:\pace.berkeleyedu Additional copies may be ordered from the berkeley orifice CPOWiNC UP IN POVERTY PROJE‘CW A Bush Center itt Child Deycloptnent 2nd ‘ i‘l I it Yale L'nn ersity. \cn llayen Oval 1 205 “LUJNH \Vavel lechnical Supplement ,4 1 WW“